Table Of Content

Wilson Bulletin, 116(2), 2004, pp. 167—171 PREDATION RISK ASSOCIATED WITH GROUP SINGING IN A NEOTROPICAL WOOD-QUAIL AMANDA M. HALE' — ABSTRACT. Decades of fruitful research on the study of vocal communication in birds have provided surprisingly little evidence ofa predation cost associated with singing. In this paper, I report the first observational evidence of a risk of predation associated with chorusing in a Neotropical wood-quail. Black-breasted Wood-Quail iOdontophorus leucolaemus) live in groups year-round and produce coordinated group choruses or duets. Three mammalian and two avian species ofpredators were attracted to playbacks ofrecorded wood-quail choruses that I used during population surveys and capture attempts from March to August, 2000-2002. The trade-offbetween signaling and predation risk may be an important force in the evolution ofchorusing in New World quails. Received 12 February 2004, accepted 19 July 2004. Long distance vocalizations are used for (Yasukawa 1989). Two recent studies provide territory defense, mate attraction, and other additional experimental support: Brown Skuas functions in a variety of animals (Bradbury (Stercororius antarcticci) use the mate attrac- and Vehrencamp 1998, Todt and Naguib tion calls of colonially nesting seabirds as a 2000). Whenever such signals are sent, how- cue for prey location and selection (Mougeot ever, there is a danger that they will be heard and Bretagnolle 2000), and long-distance call- and taken advantage ofby illegitimate receiv- ing in Crested Tits {Pants cristatiis) increases ers, such as competitors, parasites, or preda- the risk of predation by Eurasian Sparrow- tors (McGregor 1993, Bradbury and Vehren- hawks (Accipiter nisiis; Krams 2001). camp 1998, Haynes and Yeargan 1999). For In this study, I report observational evi- example, the parasitoid tachinid fly, Onnia dence that predators are attracted to group ochracea, is attracted to tape recordings of a singing in Black-breasted Wood-Quail {Oclon- host cricket, GryUiis integer (Cade 1975); the tophorus leucolaemus), a social species re- fringe-lipped bat (Trachops cirrhosiis) uses stricted to the remaining highland forest of acoustic cues to capture calling frogs (Tuttle Costa Rica. Black-breasted Wood-Quail live and Ryan 1981); and several species of fo- in coveys year-round (2 to 15 individuals) in liage-gleaning bats in Panama use the mate- the dense understory of cool, wet forest be- attracting songs of male katydids to locate tween 900 and 1,600 m elevation. Nesting these prey (Belwood and Morris 1987). In usually occurs around the onset of the rainy birds, however, evaluation of predation costs season in May and June (McDonald and Win- AMH associated with signaling has primarily been nett-Murray 1989; pers. obs.). Coveys restricted to studies of plumage conspicuous- defend group territories and produce coordi- ness (Gotmark 1992, Promislow et al. 1992, nated group choruses or duels. Neighboring Andersson 1994). Although most studies of coveys are most often heard calling back and costs associated with vocal communication in forth just after dawn. All members ofthe cov- birds have focused on energetics (Eberhardt ey participate in the chorus, and individuals 1994, Oberweger and Gollcr 2001, Thomas often hop up on perches (dead fallen trees) 2002), eavesdropping by predators has been approximately I m off the ground while sing- demonstrated experimentally in three species. ing. Depending on the weather and terrain, the TsthuedyfirsotfexRpeedr-iwmeinntgaeldevBildaecnkcbeircdosme{sAgeflraoimitas cofhor2u()s()-i5s()a(u)dimble(AtoMhHumpaenrss. forbosm.).a dBiesctaaunscee choruses and duets by a single mated pair phoenicens), in which nests accompanied by have the same basie strueture in Blaek-breasi- playback of female ‘chit' calls were depredat- ed Wood-Quail, hereafter will refer to both ed at higher rates than nests without playback I as choruses. Mi:rn()i)s ' Dept, of Biology, Univ. ot Miami. I’.O. Box 2491 IX. Coral Cables. I'l. 3.M24, U.SA: e-mail: The study area is loeated in Montevertle ahale^^bio.miami.edu (10° 15' N, 84° 46' W ). a 1,500-ha dairy farm- 167 168 THE WILSON BULLETIN • Vol. 116, No. 2, June 2004 ing community at an elevation of 1,350-1,500 determine whether wood-quail coveys adjust- m on the Pacific slope of the Cordillera de ed their responses to playbacks based on the Tilaran in Puntarenas Province, Costa Rica. number of individuals in the “intruding” cov- This area is the site of long-term ecological ey. These 16 choruses were chosen based on studies and has been well described (Nadkarni the clarity ofthe recording and the absence of and Wheelwright 2000). Monteverde is a frag- other sounds, including singing by other spe- mented landscape, composed of open pas- cies, wind, and rain. One of these tapes con- tures, some cultivated bananas and coffee, and sisted of a 15-sec chorus that was looped to considerable remnant pre-montane and lower make a 30-sec recording, while each of the montane moist/wet forest (Holdridge 1967). other tapes consisted ofa single chorus lasting The community lies on a narrow plateau 26-31.5 sec. bounded on the north by human dwellings and To simulate territorial intrusions, I entered agriculture, on the south and west by cliffs, a group’s territory and played a recording (30 and o—n the east by a large expanse of cloud sec from the cloud forest CD recording in forest the 28,000-ha Monteverde Reserve 2000 and 2001, or one of the experimental Complex, Costa Rica’s largest intact highland tapes in 2002) using a Sony Walkman, a 20- forest. Black-breasted Wood-Quail inhabit the watt RMS power amplifier, and a 12.5-cm, 15- large windbreaks and patches of remnant for- watt Speco loudspeaker (300-15,000 Hz) est in the Monteverde community and the con- placed 0-2 m above the forest floor. If the tinuous forest of the Monteverde Reserve group responded immediately, either by call- Complex. Although locally common in the ing or approaching the loudspeaker, I did not Monteverde area. Black-breasted Wood-Quail continue with playback. If the group did not are generally rare throughout Costa Rica respond, I repeated the playback once every 3 (Stiles and Skutch 1989, Fogden 1993), and min for a maximum of 15 min. Amplitude lev- are classified as a spec—ies ofconservation con- el ofthe playback was matched by ear to sing- cern by the lUCN World Conservation ing wood-quail and was held constant Union (Collar et al. 1994, Fuller et al. 2000). throughout the trials. In general, the playback In Monteverde, average covey size is four was just loud enough so that the sound trav- adults with approximately 11% of surveyed eled approximately 200 m and was not dis- coveys containing three or more adults (/z = torted. Wood-quail coveys tended to respond 151 coveys). Average density is one covey per immediately when a recording was played 3.3 ha (AMH unpubl. data). within their territories; they would sing and From March through August, 2000-2002, I approach within a few cm of the loudspeaker. conducted wood-quail surveys in the Monte- Solitary individuals also responded to play- verde area by using playbacks of choruses to backs by singing, but did not closely approach simulate territorial intrusions. A single record- the sound source. I also used playbacks to lure ing from the Indicator Birds ofthe Costa Ri- wood-quail into mist nets or large traps for can Cloud Forest CD (Ross et al. 1997) was individual marking and genetic sampling. The used in 2000 and 2001. This 58-sec recording number of times the recording was played consisted of three choruses from a bout of during capture (from 1 to approximately 20 counter-singing between two coveys of un- times per hr) depended upon the response of known size. The recordings used in 2002 were the focal group and the number of group obtained from songs provoked by the use of members captured. playback. Recordings were made from 64 RESULTS AND DISCUSSION bouts of singing by 45 coveys; I used a Sony Cassette Recorder TC-D5M and Audio- Wood-quail choruses are rushing gabbles of Technica unidirectional microphone (70- notes patterned into two sets of paired sylla- 18,000 Hz). Digital sound files were created bles that are repeated over and over. Choruses from the analogue recordings at 22.05 kHz last 10-60 sec and occur most oftenjust after with 16-bit accuracy using Syrinx Version dawn (54% of spontaneous choruses occurred 2.2b (2001). Sixteen experimental tapes, con- between 05:30 and 06:30 GMT-6, n = 87 cho- sisting of one chorus each from 16 coveys of ruses). The majority ofchoruses are answered known size (range 2-7 adults), were made to by neighboring groups, and bouts of counter- Hale • PREDATION AND GROUP SINGING IN WOOD-QUAIL 169 singing among neighbors last from about 1 approaching covey and pursued the quail on min to over 1 hr (AMH unpubl. data). When foot as they ran for cover; two Collared For- j two or more social groups come into contact est-Falcons and the Barred Forest-Falcon I near a territorial boundary, each group pro- perched on a branch located approximately 3 I duces a well-coordinated chorus and the m above the loudspeaker and visually groups will either alternate or overlap their searched the immediate area; and the dogs ran ' bouts of singing. Often, after a few bouts of directly at the loudspeaker before pausing and i counter-singing, one or more groups will re- searching the surrounding area. Two of these treat from the area. In other instances, the ter- predators are forest specialists (Collared For- ritorial encounters can last up to several hr. In est-Falcon, margay) and may be constrained addition to continued singing during territorial to using vocal cues to locate prey. Unlike the encounters, individuals produce several other other predators, the Barred Forest-Falcon is vocalizations and participate in strutting dis- too small (165-200 g; Stiles and Skutch 1989) plays, chases, and even physical fights with to capture adult wood-quail (mean = 295 g I neighboring group members. Wood-quail cov- ±16 SD, n = 103 adults; AMH unpubl. data); eys responded to simulated territorial intru- however, because these playback trials oc- sions with either choruses or approaches dur- curred during the wood-quail breeding season, ing 65% of surveys and 75% of capture at- it may have been looking for chicks. tempts in 2000-2002 {n = 467 surveys and n Although energetic costs associated with = 122 capture attempts). producing signals have received the most at- During surveys and capture attempts, I wit- tention, there is increasing evidence that ad- nessed eight instances of predator response to ditional costs, such as increased exposure to playbacks of wood-quail choruses (three in predators, should also be considered in studies 2000, four in 2001, and one in 2002). During of the evolution of communication (reviewed 109 survey mornings (mean = 4.3 social in Gil and Gahr 2002). This may be particu- groups surveyed per morning), the loudspeak- larly important in birds because the empirical er was approached by a single predator on evidence for a high energetic cost of singing four occasions: two gray foxes (Urocyon ci- in passerines is equivocal (Eberhardt 1996, nereoargenteus) and two Collared Forest-Fal- Gaunt et al. 1996, Oberweger and Goller cons {Micrastiir semitorquatus). These pred- 2001, Thomas 2002, Ward et al. 2003), and ators approached the loudspeaker after the re- studies of a nonpasserine. Red Junglefowl/do- cording had been played from one to three mestic roosters (Callus galIus), show that times. During 95 capture mornings, the loud- crowing is not energetically costly (Chappell speaker was approached on three occasions by el al. 1995, Horn et al. 1995). Similarly, cho- a single predator, one margay {Leopardus wie- rusing in Black-breasted Wood-Quail seems dii), one Barred Forest-Falcon (Micrastur ruf- unlikely to be energetically costly even icollis), and one Collared Forest-Falcon, and though it is a year-long endeavor: on any giv- on one occasion by two domestic dogs {Canis en day, individuals sing for a short amount of familiaris\ which have been observed to cap- time, perhaps giving only several bouts, each ture wood-quail in this area). While the forest- less than I min in length (AMH unpubl. data). falcons approached the loudspeaker after the This is in contrast to the high le\el of singing recording was played just three limes during effort, in terms ofboth frequency and duration capture attempts, the margay and dogs re- of songs, observed in many male passerines sponded after the recording had been played during the breeding season (Welty 1982, approximately 10 times. In each of these in- Catchpole and Slater 1995). 'Hiis comparison stances, the behavior ofthe predator suggested suggests that a high metabolic cost aiul li>ss that it was responding to the vocalizations of time that could be spent foraging or resting I rather than something else in the einironment. are not likely for wood-tjiiail. with the excep- ! The two foxes walked directly to the loud- tion of chorusing associated with prolonged I speaker before leaving the area upon seeing territorial encounters. ' me; the margay broke through the mist nets 'Hie observations jnesented here suggest I and pounced on the loutlspeaker; one C’ollared that pretlators eavestlrop on choruses pro- Forest-E'alcon laiuletl within a lew m of the duced by w()od-t|uail. In my siutly area, a ; I 170 THE WILSON BULLETIN • Vol. 1J6, No. 2, June 2004 predator responded to recordings on 1.4% of Chapman Memorial Fund, a—nd the Department of Bi- the 589 surveys or capture attempts. This is ology, University of Miami Maytag Fellowship. likely to be a great underestimate of the risk LITERATURE CITED of predation; it is reasonable to presume that I did not see all of the predators that were AndeUrnsisvoenr,sitMy.PrBe.ss,19P9r4i.ncSeetxouna,lNseewlecJteirosne.y.Princeton attracted to the playbacks, in part because they Belwood, J. J. and G. K. Morris. 1987. Bat predation avoid humans. Moreover, the suite ofdetected and its influence on calling behavior in Neotrop- predators included both those that are com- ical katydids. Science 238:64-67. monly seen (e.g.. Barred Forest-Falcon, gray Bradbury, J. W. and S. L. Vehrencamp. 1998. Prin- fox) and those that are observed only rarely ciples ofanimal communication. Sinauer, Sunder- in the Monteverde area (e.g.. Collared Forest- land, Massachusetts. Falcon, margay) (Fogden 1993, Nadkarni and Cade, W. 1975. Acoustically orienting parasitoids: fiy phonotaxis to cricket song. Science \90:\3\2- Wheelwright 2000). Detection ofthese rare or 1313. uncommon predators provides further support Catchpole, C. K. and P. J. B. Slater. 1995. Bird that the potential suite of predators is large song: biological themes and variations. Cam- and that chorusing in Black-breasted Wood- bridge University Press, Cambridge, United King- Quail may have a non-trivial cost of preda- dom. tion. Additional anecdotal evidence suggests Chappell, M. A., M. Zuk,T. H. Kwan, andT. S.John- SEN. 1995. Energy cost ofan avian vocal display: that the risk of predation associated with cho- crowingin RedJunglefowl. Animal Behaviour49: rusing will be even greater during territorial 255-257. encounters because of increased amounts of Collar, N. J., M. J. Crosby, andA. J. Stattersfield. singing and impaired vigilance. In the few 1994. Birds to watch 2: the world list of threat- cases in which I was able to closely observe ened birds. BirdLife Conservation Series, no. 4. territorial encounters, I successfully ap- BirdLife International, Cambridge, United King- m dom. proached the wood-quail to within a few Eberhardt, L. S. 1994. Oxygen consumption during without being detected. Future playback stud- singing by male Carolina Wrens {Thryothonis lu- ies should reduce variation in the amount of doviciamis). Auk 111:124-130. signal that is available to potential predators Eberhardt, L. S. 1996. Energy expenditure during from one playback trial to the next, and in- singing: a reply to Gaunt et al. Auk 113:721-723. corporate a control (i.e., recordings of silence Fogden, M. 1993. An annotated checklist ofthe birds of Monteverde and Penas Blancas. Litografia e or other sounds) to account for the likelihood Imprenta LIL, San Jose, Costa Rica. ofdetecting a predator by chance. These mod- Fuller, R. A., J. P. Carroll, and P. J. K. McGowan ifications would improve the estimate of the (Eds.). 2000. Partridges, quails, francolins, snow- frequency of eavesdropping and confirm its cocks, guineafowl, and turkeys: status survey and relative importance in this system. conservation action plan 2000-2004. World Pheasant Association, WPA/BirdLife/SSC Par- ACKNOWLEDGMENTS tridge, Quail, and Francolin Specialist Group, and World Conservation Union, Gland, Switzerland I thank the Ministerio del Ambiente y Energia and and Reading, United Kingdom. the Direccion General de Vida Silvestre Ministerio de Gaunt, A. S., T. L. Bucher, S. L. L. Gaunt, and L. Recursos Naturales, Energia. y Minas for issuing per- E Baptista. 1996. Is singing costly? Auk 113: mits inCosta Rica; the Organization forTropicalStud- 718-721. ies forlogistical support; the MonteverdeCloudLorest Gil, D. andM. Gahr. 2002. The honesty ofbirdsong: Preserve and Monteverde landowners for granting ac- multiple constraints for multiple traits. Trends in cess to their land; E. Vargas, A. Torres, and V. Ruiz Ecology & Evolution 17:133-141. for field assistance; and C. R. Hughes, W. A. Searcy, Gotmark, E 1992. Anti-predatoreffectofconspicuous D. A. Williams, I. Krams, andtwo anonymousreview- plumage in a malebird. Animal Behaviour44:51- ers for helpful comments on the manuscript. Survey 55. and capture methods were approved by the University Haynes, K. E and K. V. Yeargan. 1999. Exploitation of Miami Animal Care and Use Committee. This re- ofintraspecific communication systems: illicitsig- search was funded in part by the Organization for nalers and receivers. Annals ofthe Entomological Tropical Studies Res—earch Eellowship, the Cooper Or- Society of America 92:960-970. nithological Society Mewaldt-Kin—gStudentResearch Holdridge, L. R. 1967. Life zone ecology. Tropical Award, the University of Miami Center for Latin Science Center, San Jose, Costa Rica. American Studies Dissertation Research—Award, the Horn, A. G., M. L. Leonard, and D. M. Weary. American Museum of Natural History Frank M. 1995. Oxygen consumption during crowing by Hale • PREDATION AND GROUP SINGING IN WOOD-QUAIL 171 roosters: talk is cheap. Animal Behaviour 50: phism in birds. Proceedings ofthe Royal Society 1171-1175. ofLondon, Series B 250:143-150. Krams, I. 2001. Communication in Crested Tits and Ross, D. L., Jr., K. N. Rabenold, and T. R. Simons. the risk ofpredation. Animal Behaviour61:1065- 1997. IndicatorbirdsoftheCostaRicancloudfor- 1068. est. Library of Natural Sounds, Cornell Labora- McDonald, D. and K. Winnett-Murray. 1989. First tory of Ornithology, Ithaca, New York. reported nests of the Black-breasted Wood-Quail Stiles, F. G. and A. F. Skutch. 1989. A guide to the (Odontophorus leucolaemus). Condor 91:985- birds ofCosta Rica. Cornell University Press, Ith- 986. aca, New York. McGregor, P. K. 1993. Signalling in territorial sys- Thomas, R. J. 2002. The costs of singing in nightin- tems: a context for individual identification, rang- gales. Animal Behaviour 63:959-966. ingandeavesdropping. PhilosophicalTransactions Todt, D. and M. Naguib. 2000. Vocal interactions in of the Royal Society of London, Series B 340: birds: the use of song as a model in communica- 237-244. tion. Advances in the Study of Behavior 29:247- Mougeot, F. and V. Bretagnolle. 2000. Predation as 296. a cost of sexual communication in nocturnal sea- Tuttle, M. D. and M. J. Ryan. 1981. Bat predation birds: an experimental approach using acoustic andtheevolutionoffrogvocalizationsintheNeo- signals. Animal Behaviour 60:647-656. tropics. Science 214:677-678. Nadkarni, N. M. and N. T. Wheelwright (Eds.). Ward, S., J. R. Speakman, and P. J. B. Slater. 2003. 2000. Monteverde: ecology and conservation ofa The energy cost of song in the Canary, Serinus tropical cloud forest. Oxford University Press, Canaria. Animal Behaviour 66:893-902. New York. Welty, j. C. 1982. The life ofbirds, 3rd ed. Saunders Oberweger, K. and F. Goller. 2001. The metabolic College Publishing, Philadelphia, Pennsylvania. cost of birdsong production. Journal of Experi- Yasukawa, K. 1989. The costs and benefits ofa vocal mental Biology 204:3379-3388. signal: the nest-associated ‘Chit’ of the female Promislow, D. E. L., R. Montgomerie, and T. E. Red-winged Blackbird, Agelaiusphoeniceiis. An- Martin. 1992. Mortality costs of sexual dimor- imal Behaviour 38:866-874.