University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln

USGS Staff -- Published Research US Geological Survey

2007

Territoriality, Prospecting, and Dispersal in Cooperatively Breeding Micronesian Kingfishers (Todiramphus cinnamominus reichenbachii)

Dylan C. Kesler U.S. Geological Survey, [email protected]

Susan M. Haig U.S. Geological Survey, [email protected]

Follow this and additional works at: https://digitalcommons.unl.edu/usgsstaffpub

Kesler, Dylan C. and Haig, Susan M., "Territoriality, Prospecting, and Dispersal in Cooperatively Breeding Micronesian Kingfishers (Todiramphus cinnamominus reichenbachii)" (2007). USGS Staff -- Published Research. 664. https://digitalcommons.unl.edu/usgsstaffpub/664

This Article is brought to you for free and open access by the US Geological Survey at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in USGS Staff -- Published Research by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. The Auk 124(2):381–395, 2007 © The American Ornithologists’ Union, 2007. Printed in USA.

TERRITORIALITY, PROSPECTING, AND DISPERSAL IN COOPERATIVELY BREEDING MICRONESIAN KINGFISHERS (TODIRAMPHUS CINNAMOMINUS REICHENBACHII) Dylan C. Kesler1,2,3 and Susan M. Haig1 1U.S. Geological Survey Forest and Rangeland Ecosystem Science Center, 3200 SW Jeff erson Way, Corvallis, Oregon 97331, USA; and 2Department of Fisheries and Wildlife, Oregon State University, Corvallis, Oregon 97331, USA

Abstract.—We investigated territoriality, prospecting, and dispersal behavior in cooperatively breeding Pohnpei Micronesian Kingfi shers (Todiramphus cinnamomi- nus reichenbachii) throughout the annual cycle using radiotelemetry and color-band resights. Mean home-range size was 6.3 ha and territories were 8.1 ha. Within territories, Micronesian Kingfi shers shared 63% of their home-range space with coterritorial occupants, and 3% was shared with extraterritorial conspecifi cs. Birds on cooperative territories had larger home ranges that overlapped more with coter- ritory occupants’ home ranges than birds in pair-held territories. Despite evidence suggesting that resources necessary for survival and reproduction occurred on each territory, Micronesian Kingfi shers of all age and sex classes made extraterritorial prospecting movements. Prospecting was rare; it comprised only 4.3% of our obser- vations. When birds departed on forays, they were gone for ∼1.9 h and returned to home territories before sunset. Prospecting by dominant birds was temporally correlated with courtship and nest initiation, and birds were observed at neighbor- ing nest sites with opposite-sex conspecifi cs during the period when females were available for fertilization. Juveniles and helpers prospected throughout the year and made repeated homesteading movements to dispersal destinations before dis- persing. Mean dispersal distance for radiomarked individuals was 849 m. Results suggest that prospecting in Micronesian Kingfi shers is a complex behavior that provides information for dispersal decisions and familiarity with dispersal destina- tions. Additionally, extraterritorial movements may provide covert opportunities for reproduction, which have potential to profoundly infl uence the distribution of fi tness among helper and dominant Micronesian Kingfi shers. Received 29 August 2005, accepted 14 March 2006.

Key words: Caroline Islands, dispersal, foray, Halcyon cinnamomina, homesteading, Micronesia, prospecting, Todiramphus cinnamominus.

Territorialité, prospection et dispersion chez des Todiramphus cinnamominus reichenbachii présentant une reproduction coopérative

Rsumé.—Au cours du cycle annuel, nous avons étudié la territorialité, la prospection et le comportement de dispersion chez des Martins-chasseurs cannelle (Todiramphus cinnamominus reichenbachii) coopérant pour la reproduction. Pour ce faire, la télémétrie et la ré-observation d’oiseaux marqués de bagues colorées ont été utilisés. Le domaine vital moyen s’étendait sur 6,3 ha et les territoires, sur 8,1 ha. Les martins-chasseurs partageaient 63% de leur domaine vital avec les autres occupants du territoire et 3% avec leurs congénères extraterritoriaux. Les oiseaux occupant des

3Present address: Biological Sciences, 4095 Derring Hall, Virginia Tech, Blacksburg, Virginia 24060, USA. E-mail: [email protected] 381 382 Kesler and Haig [Auk, Vol. 124

territoires en coopération avaient des domaines vitaux plus grands qui chevauchaient davantage celui des autres occupants du territoire que les couples d’oiseaux occupant seuls le territoire. Malgré des signes évidents suggérant la présence des ressources nécessaires à la survie et la reproduction dans chaque territoire, les martins-chasseurs des deux sexes et de toutes les classes d’âge ont eff ectué des déplacements extraterritoriaux de prospection. La prospection était rare, comptant seulement pour 4,3% de nos observations. Lorsque les oiseaux partaient pour une incursion, ils s’absentaient pendant ?1,9 h et revenaient dans leurs territoires avant le coucher du soleil. La prospection par les oiseaux dominants était corrélée dans le temps avec la parade nuptiale et l’initiation des nids. Ces oiseaux étaient observés à des sites de nidifi cation voisins en compagnie de congénères du sexe opposé pendant la période où les femelles pouvaient être fertilisées. Les juvéniles et les aides au nid prospectaient tout au long de l’année. Ils ont eff ectué des déplacements répétés vers des destinations de dispersion avant la dispersion elle-même. La distance moyenne de dispersion chez les individus munis d’émett eurs était de 849 m. Les résultats suggèrent que la prospection chez les Martins-chasseurs cannelle est un comportement complexe qui les informe dans leur décisions de dispersion et leur permet de se familiariser avec les destinations de dispersion. De plus, les mouvements extraterritoriaux peuvent fournir des opportunités clandestines pour la reproduction, lesquelles ont le potentiel d’infl uencer profondément la distribution du fi tness parmi les martins-chasseurs dominants et les aides au nid.

Knowledge of dispersal and space use is p rospecting movements into the surrounding essential for testing hypotheses about the evolu- landscape (Bowen et al. 1989, Reed et al. 1999, tion of sociality in cooperatively breeding spe- Koenig et al. 2000, Fedy and Stutchbury 2004). cies. Predominant theories about cooperative However critical this information is to under- breeding suggest that there are fi tness benefi ts standing sociality in cooperative breeders, associated with delayed dispersal (“benefi ts of prospecting movements are rare and elusive, philopatry” hypothesis; Stacey and Ligon 1991), and att empts at empirical assessments can be including the potential to inherit resources beset by methodological biases (Koenig et al. when natal or nearby territory vacancies occur. 1992, 2000; Walters 2000; Clobert et al. 2001; The cost of delaying can be outweighed if the Johnson and Horvitz 2005). Thus, few have inherited resources provide enhanced fi tness identifi ed how prospecting relates to disper- when compared with those in outlying areas sal, where prospecting birds travel, and what (Komdeur 1991, 1992, 1994). Timing of dispersal information birds acquire while prospecting, is variable among species, however, and litt le is despite the paramount importance of these data known about the proximate factors infl uencing to understanding proximate dispersal decisions when and how dispersal occurs. in cooperative breeders (Walters 2000; but see In many cooperatively breeding species, Doolan and Macdonald 1996, Schjørring et al. some individuals disperse as juveniles, whereas 1999, Fedy and Stutchbury 2004). others delay for extended periods (Koenig and Once delayed dispersal has arisen, the behav- Pitelka 1981, Brown 1987, Stacey and Koenig ior can be maintained by extrinsic limitations 1990, Walters et al. 1992, Ligon 1999, Ekman et in resources necessary for reproduction (“eco- al. 2004). This patt ern implies that each individ- logical constraints” hypothesis; Emlen 1982). ual gathers information about resource quality, Habitats suitable for survival and nesting of territory occupancy, and breeding vacancies in cooperative species are oft en saturated with ter- nearby areas before making dispersal deci- ritories that are packed boundary-to-boundary sions. Some have proposed that birds obtain and aggressively defended, which suggests information pertinent to dispersal through that territory availability may function as the social interactions along territorial boundaries limited resource that prevents young from (Hale et al. 2003) and by making extraterritorial dispersing (“habitat saturation” hypothesis; April 2007] Movement and Space Use in Micronesian Kingfi shers 383

Selander 1964, Brown 1974, Gaston 1978, Stacey endangered under the U.S. Endangered Species 1979, Koenig and Pitelka 1981). In pair-breeding Act following precipitous declines caused by species, investigators have identifi ed spatially introduced brown tree snakes (Boiga irregularis; and temporally variable factors that infl uence U.S. Fish and Wildlife Service 1984, Savidge the distribution of individuals, including envi- 1987). They now exist only as a captive popula- ronmental conditions, intraspecifi c interactions tion in U.S. zoos (Haig and Ballou 1995, Haig (Brown 1964, Emlen and Oring 1977, Hixon et al. 1995, Bahner et al. 1998, Kesler and Haig 1980, Schoener 1983, Lima 1984, Stamps and 2004). Plans have been announced for reintro- Krishnan 1999), and resource dispersion and ducing the species back into its last native habi- predictability (Brown 1964, Emlen and Oring tats on Guam, but this cannot be done without 1977, Bollmann et al. 1997, Clark and Shutler information about spatial distribution and dis- 1999, Zwicker and Walters 1999, Tyre et al. persal (U.S. Fish and Wildlife Service 2004). 2001). Yet few have investigated proximate mechanisms underlying space use and ter- Metods ritoriality in cooperative species (Langen and Vehrencamp 1998, Breininger and Oddy 2004, Research was conducted on the island Fedy and Stutchbury 2004). of Pohnpei, Federated States of Micronesia Within a territory, each individual bird uses (6°52’N, 158°13’E). Pohnpei is a circular island space, which makes up its home range (White ∼20 km in diameter and it has the highest peak and Garrott 1990), and together the home ranges in the Micronesian chain (∼800 m; Engbring et for a pair or social group comprise a territory. al. 1990). Extensive lowland coastal plateau and Space within territories is aggressively defended mangrove swamps surround the inner moun- from conspecifi cs along territory boundaries tain range, which is characterized by dense (Brown 1964). There is a lack of information tropical rainforests. Three sites were selected for about how space is partitioned among indi- this investigation: the Ranch (6°57’N, 158°12’E), viduals, and the actual extent to which spatial College of Micronesia (6°54’N, 158°9’E), and resources are distributed among a cooperatively Palikir study areas (6°55’N, 158°9’E). Each site breeding group has never been assessed, to our included strand vegetation, early-succession knowledge. Although birds within a territory and mature lowland rainforest, grassland, urban are generally believed to share resources, some vegetation, and agroforest. Grassland included space may also be reserved for specifi c social pastures and fallow fi elds. Early-succession classes, and relationships may diff er among rainforest and agroforest vegetation were char- cooperative and pair-held territories. acterized by lower canopy (2–20 m high) hibis- We investigated territoriality, prospect- cus (Hibiscus tiliaceus), banana (Musa sapientum), ing movements, and dispersal behaviors in coconut (Cocos nucifera), breadfruit (Artocarpus cooperatively breeding Pohnpei Micronesian altilis), and sakau (Piper methysticum). Mature Kingfi shers (Todiramphus cinnamominus reichen- forests had higher canopies (25–30 m high) and bachii). We synthesized observations of color- were dominated by mango (Mangifera indica), banded and radiomarked birds to assess the dohng (Campnosperma brevipetiolata), sadak distribution and movement of individuals (Elaeocarpus carolinensis), karara (Myristica within and among territories. We evaluated insularis), ais (Parinari laurina), and tree ferns home ranges, territories, prospecting distances, (Cyathea spp.; see Mueller-Dombois and Fosberg timing, and the relationship between disper- 1998, Buden 2000). sal and extraterritorial movements using a Study population.—We follow the terminology global information system (GIS) and behavior of recent literature (e.g., Haydock and Koenig observations. Additionally, radiotelemetry and 2002) by using “dominant” to refer to the puta- color-band observations were used to assess tive breeders on a territory and “helper” for within- and among-year dispersal. Results from off spring that have delayed dispersal through the present study are intended to lend insight subsequent parental reproductive att empts. into space use and dispersal and provide Additionally, we use “juvenile” to describe prog- information for recovery eff orts for the Guam eny from the most recent breeding season. We Micronesian Kingfi sher (T. c. cinnamominus). radiomarked and color-banded a population of The Guam Micronesian Kingfi sher was listed as Micronesian Kingfi shers on study areas between 384 Kesler and Haig [Auk, Vol. 124

January 1999 and November 2004. The study excluded (n = 17). Most birds were located once populations were intensively observed from daily, and all observations were separated by January to July 1999, March to September 2000, ≥2 h to avoid autocorrelation (mean temporal September 2001, September 2002 to January 2003, separation of subsequent observations = 27 October and November 2003, and September h). Additionally, kernel density analyses can 2004. Birds were radiomarked during 1999, 2000, be biased by clusters of locations during the and 2002, and color-banded every year. Because breeding season (White and Garrott 1990), so observers were present on study areas daily, we we eliminated 104 locations that were within believe that the social classes (dominant, helper, 15 m of nest sites. The process yielded 3,929 juvenile) of all individuals were known. Birds locations. were captured in mist nets and fi tt ed with a 1.8-g Diff erential detection probabilities and miss- telemetry package (Holohil Systems, Ott awa, ing data have the potential to bias representa- Canada) using the leg-harness design (Rappole tions of movement and resource use (Porter and Tipton 1991) and a unique combination of and Dooley 1993; Koenig et al. 1996, 2000). For colored leg bands. We observed no negative example, data are biased when observers fail to eff ects from capture, observation, color bands, identify the presence or location of color-banded or radiomarking. Age and sex of each bird was or radiomarked animals that have moved off determined using molecular analyses and plum- study areas or into portions of the landscape that age characteristics (D. C. Kesler et al. unpubl. hinder detection. We avoided biasing results by data). In Pohnpei, adult Micronesian Kingfi shers locating birds every time we att empted to fi nd older than one year have white breast plumage them, with the exception of several att empts cut (Pratt et al. 1987, Fry and Fry 1992), whereas short by factors unrelated to bird movement, fl edglings and juveniles have varying degrees such as weather and equipment failures. Several of rufous breast plumage. The rufous feathers study-specifi c factors allowed us to consistently progressively molt into white along the ventral locate birds, including presampling identifi ca- feather tract (D. C. Kesler et al. unpubl. data). tion of biasing factors, the 1- to 2-km detection The plumage of helpers occasionally shows distance of the Holohil transmitt ers in the gener- remnant rufous coloration, but plumage of older ally fl at landscape, and the short-distance move- helpers is similar to that of dominants (D. C. ments of Micronesian Kingfi shers. When birds Kesler pers. obs.). were not immediately identifi ed on their home Radiotelemetry and home range.—We used territory, observers traversed the surrounding radiotelemetry and GIS analyses to evaluate landscape until the location of prospecting indi- territoriality and movement in Micronesian viduals was identifi ed. Radiotelemetry results Kingfi shers. Hand-held Yagi antennas, com- can also be biased if diurnal movement patt erns passes, and global positioning systems (GPS; correlate with telemetry sampling regimes. March III, Corvallis Microtechnologies, Thus, we repeatedly sampled throughout the Corvallis, Oregon) were used to record bearing day and night. We focused most sampling on groups of two to fi ve (mean = 2.98) directional daylight hours, because observer comments bearings for each Micronesian Kingfi sher (n = and 60 sequential telemetry locations recorded 2,108 locations). If birds were visually observed during nightt ime hours indicated that birds during telemetry sessions, we recorded their do not move substantially between sunset and actual location using the off set function of the sunrise. Additionally, we balanced sampling GPS (n = 1,942 locations). Bearing groups were equally during each 2-h time block between used to estimate the best maximum-likelihood 0600 and 1800 hours (approximate sunrise and locations for each bird using default sett ings in sunset at 7°N latitude; observation [mean ± LOAS (Ecological Soft ware Solutions, Urnäsch, SD] per time block = 649 ± 69). During telem- Switzerland; see Acknowledgments), and they etry observation periods, behaviors were also were excluded if LOAS failed to identify a recorded if the observers could visually identify location or if observers noted that birds moved individuals. Particular emphasis was placed on during the observation period. Locations recording aggressive interactions, fi ghts, and with 95% error ellipses (White and Garrott courtship behaviors, and noting behavior dur- 1990) >5,000 m2 (∼10% of the area of a mean ing prospecting movements. We believe this Micronesian Kingfi sher home range) were also method yielded a data set that was temporally April 2007] Movement and Space Use in Micronesian Kingfi shers 385 and spatially representative of Micronesian the linear-model analysis of variance (ANOVA; Kingfi sher space use and movement throughout Proc GLM) function of SAS ANALYST was used the annual cycle. to evaluate social class comparisons. Chi-square Prospecting, dispersal, and behavior.—We analyses were calculated by hand. Statistical employed a combination of empirical and tests used for each analysis, adjustment meth- subjective criteria to identify telemetry loca- ods for multiple comparisons, 95% confi dence tions recorded while birds were prospecting. intervals (95% CI), and least-squared mean First, we used the ARCVIEW animal move- parameter estimates are presented whenever ment extension (Hooge and Eichenlaub 1997) appropriate. Unless otherwise noted, estimates to conduct kernel density analyses (White and are reported as means ± SD, and diff erences are Garrott 1990) of telemetry locations. Home considered statistically signifi cant at α < 0.05. range was considered to be the 95% use contour of each bird within a particular year (White Results and Garrott 1990, Roshier and Reid 2003). The 95% use contours for dominants on each terri- From 1999 to 2004, 57 male and 53 female tory were then amalgamated to delineate the Micronesian Kingfi shers were marked with boundaries of 16 focal study territories (see individual-specifi c color-band combinations. Kesler and Haig 2005b). Kernel polygons were On 16 focal territories, 43 birds were marked as considered to be outliers and excluded from the dominants, 12 as helpers, and 39 as juveniles or territory amalgamations if they were distinctly nestlings. Fift y-seven were observed during only separated from the main territory polygon the year they were marked, 22 were observed and surrounded only a single telemetry point. during two fi eld seasons, 13 were observed for Extraterritorial movements were then defi ned as three seasons, one bird was observed during locations of birds outside territorial boundaries a fi ft h, and one during a sixth season. During by >50 m, a fi gure selected because it represents the 1999, 2000, and 2002 breeding seasons, 54 approximately half the radius of a mean-sized Micronesian Kingfi shers were radiomarked circular home range. Radiomarked birds were and tracked for ∼16 weeks each (Tables 1 and considered to have dispersed if they remained 2). Additionally, one female and two males were at a prospecting destination for >24 h and aft er radiomarked during consecutive years, yielding daily movements to home territories ceased. a total of 57 kingfi sher*radio years. We considered that color-banded individuals Space distribution within and among territories.— dispersed if they moved from one territory to Home ranges were evaluated for radiomarked another between years. Micronesian Kingfi shers (Table 1). The mean Analysis.—Statistical analyses were con- error ellipse for bearing groups was 409 m2, and ducted using SAS, version 8e (SAS Institute, a mean of 64 ± 22 locations (minimum 13, maxi- Cary, North Carolina). Two-sample t-tests were mum 100) were used for each home-range esti- used to make comparisons between sexes, and mate. Mean home-range size was 7.31 ± 6.83 ha

Table 1. Estimates of home-range size and prospecting distances from home territories for each sex and social class of Micronesian Kingfi shers during 1999, 2000, and 2002, using radiotelemetry and kernel-density analyses. Estimates (mean ± SD) are presented for each class.

All observations Prospecting Mean Home-range Observa- Mean Sex Class Birds a observations size (ha) b Birds tions (%) distance F Dominant 16 69 ± 20 5.7 ± 2.7 6 0.7 77 F Helper 3 63 ± 23 7.8 ± 4.8 2 9.0 399 F Juvenile 9 56 ± 13 5.2 ± 3.7 8 9.4 221 M Dominant 16 67 ± 18 7.1 ± 2.7 9 1.2 114 M Helper 7 60 ± 33 7.9 ± 4.4 6 11.0 429 M Juvenile 6 51 ± 27 4.6 ± 3.4 1 2.8 189 a Sample sizes are presented in bird*years. b Data for three dispersing individuals are addressed elsewhere, because they exerted undue leverage on estimates. 386 Kesler and Haig [Auk, Vol. 124

Table 2. Home-range overlap within and among Micronesian Kingfi sher territories on Pohnpei during 1999, 2000, and 2002. Results are presented as area (ha), with sample sizes in parentheses. P-values are reported for two-sample t-tests for equal means.

Overlapping home range area Overlap type Neighbors Within territories P-value Dominant–dominant 0.14 (22) 4.39 (8) <0.0001 Dominant–helper 0.34 (12) 4.78 (10) <0.0001 Dominant–fl edgling 0.11 (21) 3.29 (9) <0.0001 Fledgling–helper 0.04 (2) 4.74 (3) 0.017 Helper–helper a 1.18 (6) – – Fledgling–fl edgling 0.01 (20) 2.05 (4) <0.0001 a Home ranges of neighboring helpers overlapped signifi cantly more than home ranges of all other neighboring social classes. for all radiomarked birds. The disparate point radiomarked birds, or the amount of home- distributions of three individuals that dispersed range overlap (Table 2). Forty-two birds were during observations yielded estimates that radiomarked while a coterritorial occupant was exerted undue leverage on results, so they were simultaneously tracked, and dyad combina- excluded and are treated below. Aft er disperser tions of these individuals yielded 34 pairs of data were removed, mean home-range size was simultaneous coterritorial home ranges. Home 6.28 ± 3.33 ha (Table 1). ranges of coterritorial birds overlapped by a There was suggestive evidence that home- mean of 3.97 ± 2.10 ha. No diff erences were range sizes diff ered among social classes identifi ed between the area overlapped by two (ANOVA, P = 0.09). Juvenile home ranges were dominants, dominants and helpers, dominants the smallest and helper home ranges were and juveniles, helpers and juveniles, or two the largest, but pairwise comparisons were juveniles (one-way ANOVA, F = 1.78, df = 4 and insignifi cant (Scheff e’s adjustment for multiple 33, P = 0.16; Table 2). In regard to mean home- comparisons, P > 0.05, 95% CI: –0.36 to 6.23 ha; range size, results illustrate that birds share 63% Table 1). Similarly, there was no diff erence in the of their home range with other individuals on home-range size of males and females (t-test; P = the same territory, but that not all space is com- 0.26). The relationship between sociality and monly used by all territory occupants. When home-range size was evaluated by comparing a comparison was made between mean home- home ranges of birds in cooperative group-held range overlap on cooperative (mean ± SE, 2.75 ± territories (n = 31) with those on pair-held terri- 0.39 ha, n = 12) and pair-held territories (4.63 ± tories (n = 26). Overall, the home ranges of birds 0.46 ha, n = 22), results indicated that home on pair-held territories were 1.8 ha smaller than ranges on cooperative territories overlapped by those on cooperative territories (t-test, P = 0.04, 1.88 ha more than those on less-crowded pair- 95% CI: 0.1–3.5 ha). By social class, dominant held territories (two-sample t-test, P = 0.01, 95% home ranges were 2.4 ha larger (P = 0.009; 95% CI: 0.48–3.29 ha). CI: 0.7–4.0 ha), and the home ranges of fl edglings Micronesian Kingfi sher occupancy on 16 focal were 1.8 ha larger (P = 0.04; 95% CI: 0.1–3.5 ha) on territories was recorded between 1999 and 2003. cooperative territories. Territories were defended (in territo ry*years) by Groups or pairs of Micronesian Kingfi shers single individuals (n = 2), breeding pairs (n = 13), share space within each territory. However, a single dominant and a juvenile (n = 1), breed- aggressive interactions among coterritorial ing pairs and juveniles (n = 25), and cooperative occupants suggest that not all space is com- family groups (n = 24). Dominant mortalities monly used by every individual, and that por- altered group membership in 4 territory*years, tions of each territory may be reserved for use and social composition could not be determined by only specifi c birds. To gain insight into the for 11 territory*years. When home ranges of sharing of space within territories, we assessed radiomarked dominants were amalgamated the amount of space used simultaneously by to empirically defi ne each territory, the mean April 2007] Movement and Space Use in Micronesian Kingfi shers 387

territory size was 8.1 ha. There was no diff erence were observed prospecting. The diff erence in in the size of territories that hosted cooperative means yields an estimated foray duration of groups during at least one fi eld season (n = 12), 1.9 h (t-test, P = 0.0012, 95% CI: 0.7–3.1). We also and consistently pair-held territories (n = 4) (F = observed round-trips on four occasions, when 2.41, df = 1 and 14, P = 0.14). birds were detected on their home territory, When conspecifi cs were encountered within while prospecting, and then back on their home territories or along territory boundaries, the territory again on the same day. The mean esti- birds were excluded through aggressive inter- mated departure time, prospecting duration, actions. Chases and bill-swooping displays and return time for round trips were similar to were observed 99 times during the course of those identifi ed above (0935 hours, 3.0 h, and fi eldwork. Among the aggressive displays 1237 hours, respectively). observed on the study areas, 29 were located Prospecting frequency diff ered among near territorial boundaries delineated by telem- social classes (ANOVA, arcsine-square-root- etry, and 13 were inside territories. All age and transformed proportion observations pros- sex classes participated in the behaviors, which pecting, F = 4.32, df = 2 and 61, P = 0.018), usually terminated when one or more territory with prospecting comprising 2.4% more of occupants chased intruders from confrontation the helper telemetry observations than those areas. We evaluated the eff ects of spatial exclu- of the dominants (Bonferroni correction and sion by assessing home-range overlap among 83 back-transformed, P = 0.03, 95% CI: 0.01–8.5%). dyad pairs of radiomarked birds on neighbor- Similarly, helpers prospected 161 m farther ing territories. The home ranges of neighboring from home territory boundaries than dominants birds overlapped less than those of birds on (ANOVA, F = 4.95, df = 2 and 29, P = 0.014, 95% the same territory (0.21 vs. 3.97 ha; two-sample CI: 31–292 m). No diff erences were identifi ed t-test, P < 0.001), a patt ern that applied to all between juveniles and dominants or juveniles combinations of social classes (Table 2). The and helpers (P > 0.05). Sexes did not diff er (t-test, home ranges of the six neighboring pairs of P > 0.05), and no diff erence was identifi ed among helpers overlapped more than other combina- cooperative and pair-held territories (t-test of tions of neighboring dominants, helpers, or arcsine-square-root-transformed proportion juveniles (ANOVA, F = 12.96, df = 5 and 77, P < observations prospecting, P > 0.05; mixed-models 0.0001), which indicates that helpers invaded ANOVA, P > 0.05). Although restricted samples the space of extraterritorial conspecifi cs more prevent analysis of sex by social-class interac- than other social classes. tions, the patt ern of prospecting frequencies sug- Prospecting movements.—Prospecting was gests similarity between juvenile and dominant observed in 31 individuals of all age and sex males and between juvenile and helper females. classes (Table 1). Of 3,929 telemetry locations We evaluated the temporal distribution recorded during our investigation, 170 were of prospecting by comparing the number of observed during 152 prospecting movements observed monthly prospecting movements with (multiple locations were recorded during long the number expected if they were proportional forays). Observer remarks about behavior to all telemetry observations. Prospecting move- accompanied 113 of the prospecting move- ments were temporally disproportionate to pre- ments, and the GIS verifi ed that all locations dictions for all three social classes (dominants: were >50 m from home-territory boundaries. χ2 = 16, df = 2, P < 0.001; helpers: χ2 = 54, df = 7, All prospecting movements terminated with P < 0.001; juveniles: χ2 = 21, df = 4, P < 0.001; cells birds returning to home territories before sun- with predicted occurrence <5 were excluded). set. We used two methods to evaluate mean Additionally, the monthly proportion of pros- departure time, mean foray duration, and mean pecting observations diff ered among dominants, return time. On 59 occasions, we observed birds helpers, and juveniles (Kruskal-Wallis test, χ2 = on their home territories and prospecting dur- 6.86, df = 2, P = 0.029). The timing of dominant ing the same day. We estimated a mean depar- prospecting corresponded with reproduction, ture time of 1025 hours (1.7 SE, n = 27) and a because the proportion that occurred during mean return time of 1221 hours (2.4 SE, n = 27) breeding months (April–August) diff ered from by averaging the times birds were observed that during nonbreeding months (t-test, F = on their home territories with the times they 10.2, df = 1 and 9, P = 0.01; estimated 2.3% more 388 Kesler and Haig [Auk, Vol. 124

Fig. 1. Temporal distribution of observed courtship behaviors and prospecting movements among social classes of Pohnpei Micronesian Kingfishers. Crosshatching denotes period of nest initiations on focal territories between 1999 and 2004. Courtship observations included copula- tions, courtship feedings, and nest excavations (n = 7, 6, and 43, respectively) between 1999 and 2004. Histograms illustrate the monthly proportion of courtship behaviors among all behaviors, and monthly proportion of extraterritorial observations, for radiomarked Micronesian Kingfishers in each social class during 1999, 2000, and 2001.

p rospecting observations during the breeding that the neighboring female was present on at season, 95% CI: 0.7–3.9%; Fig. 1). Although the least one of those occasions while her mate was histogram in Figure 1 appears to illustrate a simi- elsewhere. Furthermore, estimates based on lar patt ern between courtship behavior and pros- hatch dates for the nests visited by prospect- pecting in helpers, mean monthly prospecting ing males (n = 3) indicated that visits occurred movements did not diff er signifi cantly between within three days of respective laying dates, the breeding and nonbreeding months for help- when females were presumably available for ers or juveniles (t-test, P > 0.05 for both). fertilization. The visited females then went on Behavioral observations support asser- to nest with their mates on home territories. tions that extraterritorial movements may be Courtship calling with nonmates was observed motivated by birds’ search for extrapair mat- during four extraterritorial movements. ing opportunities. Courtship behaviors were Aggressive territorial head-dipping displays, observed in association with eight forays. Male aerial pursuit chases, and territory calling also helpers were observed at the nests of neighbor- characterized prospecting observations (n = 2, 5, ing birds three times, and telemetry showed and 9, respectively). April 2007] Movement and Space Use in Micronesian Kingfi shers 389

Philopatry and dispersal.—Five birds color- died, and the other bred with a new mate the banded as juveniles (one male, four females) following year. Inferences about territory own- and fi ve banded as helpers (four males, one ership can also be drawn from changes in terri- female) were observed on natal territories dur- tory occupancy across years. New mates on two ing subsequent years. However, none remained territories replaced dominant males, whereas to obtain new mates and breed on natal areas, the dominant females remained across breed- which suggests that territory inheritance by ing seasons. On six territories, dominant males helpers or off spring is rare or does not occur in remained, whereas female mates were replaced. Micronesian Kingfi shers. Three males and fi ve females dispersed from natal territories during Discussion our investigation. Two (one male, one female) dispersed as juveniles, four (one male, three Pohnpei Micronesian Kingfi shers hold all- females) remained as helpers for one breeding purpose territories throughout the year. Their season before dispersal, and two (one male, one high degree of territoriality was illustrated by female) were not observed between fl edging results showing that the home ranges of coter- and fi lling a vacancy on neighboring territories ritorial occupants overlapped by 63%, whereas two years later. Three birds dispersed while the home ranges of birds on neighboring ter- being radiotracked (one male: 1,023 m; two ritories overlapped by only 3%. Brown (1964) females: 849 and 716 m), and the remaining fi ve suggested that territoriality would evolve if color-banded birds dispersed between fi eld sea- resources were distributed so that an individual s ons (two males: 419 and 1,372 m; three females: could satisfy its nesting requirements, food sup- 295, 419, and 171 m). No diff erences were found ply, and att raction of a mate in a relatively fi xed between mean dispersal distances of radio- area, and if individuals could balance the costs marked and color-banded birds (t-test, P = 0.31, of defensive aggression with the benefi ts of n = 8), nor between males and females (t-test, defended resources. Accordingly, Micronesian P = 0.14, n = 8). Dispersal had been previously Kingfi shers are terrestrial generalists that pri- evaluated in terms of the number of territories marily subsist on the abundant invertebrate and between natal areas and breeding locations, and lizard prey items (orders Isoptera, Lepidoptera, our observations show a 2.05 territory-width and Othoptera; Emoia spp; Family Gekkonidae; separation between natal and breeding terri- D. C. Kesler unpubl. data). Birds nest in arbo- tories in radiomarked Micronesian Kingfi shers real termitaria that are apparently not limited (assuming a diameter of 320 m from circular in abundance (Kesler 2002, Kesler and Haig territory with mean territory area). Before dis- 2005b), and they do not require specialized cover persal, three radiomarked individuals made resources in the amiable climactic conditions on repeated prospecting visits (mean = 13) to the Pohnpei (Kesler and Haig 2005a). Additionally, locations where they eventually dispersed. The <1% of our telemetry locations from dominant fate of 31 color-banded juveniles and 10 help- kingfi shers were extraterritorial, which further ers that disappeared from study areas remains suggests that Pohnpei Micronesian Kingfi sher unknown, but many of these individuals prob- territories contained all the resources necessary ably dispersed to surrounding areas where they for survival and reproduction. were not subsequently resighted. On cooperative territories, home ranges Insight into territory ownership and inheri- of dominants were larger than those on pair- tance comes from the behavior of territory- held territories. Resource availability has been holding dominants following death of their shown to cause sociality in Seychelles Warblers mates. Territory ownership is not reserved for (Acrocephalus sechellensis; Komdeur 1991, 1992, either sex. A dominant male was left as the single 1994) and Red-cockaded Woodpeckers (Picoides territory holder when a hunter shot his radio- borealis; Walters et al. 1992), and larger territo- marked mate, and he remained on the territory ries and greater resources have been correlated with the previous year’s off spring. On two other with sociality in other cooperative species, such territories, dominant females retained owner- as Splendid Fairy-wren (Malurus splendens; ship aft er the deaths of radiomarked mates. One Brooker and Rowley 1995). Greater or higher- of these females was observed courtship-calling quality resources on natal areas can also infl u- with an unmarked male 17 days aft er her mate ence reproductive success (Forbes et al. 2002, 390 Kesler and Haig [Auk, Vol. 124

Luck 2003, Lõhmus and Väli 2004) and breeding envisaged for cooperative breeders (Reed et al. behavior (Emlen and Oring 1977, Walters et al. 1999), and simulations have shown the relative 1992, Byrkjedal et al. 1997, Pribil and Searcy benefi ts of such a “foray search” strategy over 2001), which may induce potential dispersers the random strategies (Boulinier and Danchin to delay. Thus, additional insight into restric- 1997, Conradt et al. 2003); many male Red- tions placed on potential dispersers by resource cockaded Woodpeckers exemplify the patt ern availability on the landscape, as well as the rela- (Walters et al. 2004). None of the radiomarked tive resource benefi ts that might be obtained on Micronesian Kingfi shers employed a fl oater dis- natal areas by delaying, would be gained from persal strategy during 3,929 telemetry*bird*days. additional study of the interaction between Instead, birds made movements that resembled specifi c foraging and nesting resources, repro- a stay-and-foray strategy, because prospecting ductive success, and cooperative breeding in was directed, repeated, and short in duration Micronesian Kingfi shers. and birds returned to their home territories Prospecting.—Juvenile and helper Micronesian before sunset. This may refl ect att empts to retain Kingfi shers may prospect to gather informa- social status and nepotistic benefi ts on a natal tion about potential dispersal destinations. territory (Ekman et al. 2001, 2004; Dickinson and In the cooperatively breeding Red-cockaded Hatchwell 2004) while simultaneously gathering Woodpecker, there is evidence that reduced contemporary information about extrinsic condi- familiarity with the environment decreases tions with which to make informed decisions disperser fi tness (Pasinelli et al. 2004). Previous about the costs and benefi ts of dispersal. investigations have also shown that knowledge Our results suggest that prospecting may pro- of localized resources can aff ect foraging effi - vide an opportunity to solicit reproduction (e.g., ciency, territoriality, predator detection, and mate Pitcher and Stutchbury 2000). Although parent- att raction (Greenwood 1980, Greenwood and age has not been investigated in Micronesian Harvey 1982, Pärt 1994, Smith and Metcalfe 1997, Kingfi shers, previous studies of other coop- Bensch et al. 1998). For Micronesian Kingfi shers, eratively breeding species show high levels of nesting and foraging resources are vital to daily promiscuity and extrapair paternity (Mulder et survival and reproduction, and experiences dur- al. 1994, DeLay et al. 1996, Li and Brown 2000, ing prospecting may educate birds about the rel- Richardson et al. 2001; but see Haig et al. 1994). ative availability of these resources on natal areas Prospecting movements of dominants were and in the surrounding landscape. We observed temporally aligned with courtship observations birds foraging, excavating nest cavities, and and nest initiations (Fig. 1), and we observed in proximity to the nests of neighbors while prospecting individuals meeting neighboring prospecting. A similar nest-resource-assessment females at nest sites several days before lay- behavior has been observed in other cooperative ing. In contrast to previous descriptions of other species (e.g., Greeno Wo dhoopoes [Phoeniculus species (Reed et al. 1999), some prospecting purpureus]; Ligon and Ligon 1990) and in many Micronesian Kingfi shers were not failed breeders pair-breeding species (see Reed et al. 1999). that were merely gathering information about While prospecting, Micronesian Kingfi shers may resources, because they went on to breed with have also been assessing the ability of conspecif- mates on home territories shortly thereaft er. ics to defend territories through direct interac- Reproduction obtained during extraterrito- tions like territorial head-dipping displays, rial movements has the potential to greatly aerial pursuit chases, and territory calling, or enhance the fi tness of prospecting birds while through indirect indicators like plumage (e.g., simultaneously reducing that of cuckolded Fan-tailed Widowbird [Euplectes axillaris]; Pryke males. Additionally, costs to prospectors may and Andersson 2003). be small, because we observed no mortality Investigators have suggested that in coopera- during extraterritorial movements. Covert tive species, potential dispersers choose either reproduction by delayed dispersers during to remain on a natal territory as a helper and forays would provide evidence of a previously “stay-and-foray” or depart from natal territories undocumented pathway to fi tness, and suggests and search as a “fl oater” until they fi nd a terri- that delaying and foraying may be more of an tory vacancy to fi ll (Brown 1987; Walters et al. alternative life-history strategy than simply 1992, 2004). Stay-and-foray models have been “making the best of a bad situation,” as some April 2007] Movement and Space Use in Micronesian Kingfi shers 391 have suggested (e.g., Emlen 1997). Further, if making dispersal decisions. Prospecting move- covert extraterritorial helper reproduction is ments of dominant birds were temporally asso- common in many species, the behavior even ciated with courtship and nesting activities, and has the potential to alter perceptions about the birds were observed in proximity to opposite- costs and benefi ts of cooperative breeding and sex neighbors, which suggests that prospecting delayed dispersal, which have been debated may also be used to solicit reproduction. for several decades (Skutch 1935, Brown et al. Results from this investigation suggest that 1978, Vehrencamp 1980, Brown 1987, Stacey a reintroduced population of Micronesian and Koenig 1990, Gerlach and Bartmann 2002, Kingfi shers on Guam would require at least Haydock and Koenig 2003, Griffi n et al. 2003). enough space for birds to maintain territories A molecular genetic investigation of parentage ∼8.1 ha in size. The limited prospecting and dis- in Micronesian Kingfi shers has the potential to persal distances detected here also suggest that lend insight into this phenomenon. a recovering population of Guam Kingfi shers Dispersal.—Although rare, long-distance dis- should not be expected to disperse across the persal has been observed in other cooperatively landscape quickly. Additional investigation into breeding species (Bowen et al. 1989, Koenig parentage, and the interaction between specifi c et al. 1996). During our study, Micronesian resources, dispersal decisions, and population Kingfi shers neither prospected nor dispersed demography, would lend further insight into great distances from home territories. The island the costs and benefi ts of cooperative breeding in of Pohnpei is limited in size (∼20 km in diam- Micronesian Kingfi shers. eter), so dispersal distances may be restricted by behavior and geography. Acnowledgments An increased probability of obtaining a high- quality breeding territory by delaying dispersal We are indebted to all who have provided is fundamental to the benefi ts-of-philopatry fi nancial and technical support to this proj- hypothesis (Stacey and Ligon 1991). By delay- ect. The U.S. Geological Survey Forest and ing, individuals can inherit natal areas or occupy Rangeland Ecosystem Science Center, U.S. neighboring territories when vacancies occur. Fish and Wildlife Service, Conservation Thus, inheritance is characteristic of many coop- Endowment Fund of the American Zoo and eratively breeding species (Hale et al. 2003), and Aquarium Association, National Geographic some investigators have even suggested that Society, Disney’s Animal Kingdom, St. Louis long-term data sets might yield evidence of Zoo Field Research for Conservation Fund, dynasties (Emlen 1997). To the contrary, our fi ve- Brookfi eld Zoo, Friends of the National Zoo, the year investigation of Micronesian Kingfi shers Micronesian Kingfi sher Species Survival Fund, provides no evidence of helpers obtaining breed- and the Milwaukee County Zoo have provided ing vacancies through inheritance, because none fi nancial backing for the project. B. Dugger, of the delayed dispersers bred on natal territo- L. Houck, L. Kesler, I. Lopes, O. Taft , and J. ries. However, some helpers dispersed to neigh- Walters provided valuable comments on study boring territories, where they bred in subsequent design and manuscript development. D. Buden years, so individuals may be queuing for breed- and B. Raynor provided advice on Pohnpei. ing vacancies in nearby areas (Zack 1990). We also extend thanks to the government of In summary, the Micronesian Kingfi sher is the Federated States of Micronesia, H. Anson, a highly territorial species that maintains all- R. Mauricio, the landowners in Pohnpei, and purpose, year-round territories as pairs and everyone in the Haig Lab. M. Boris, C. Cantero, cooperative groups. No Micronesian Kingfi shers J. Santos, and P. Sanzenbacher also made became fl oaters during our investigation, but all contributions as fi eld assistants and cultural age and sex classes made short-duration pros- guides. For information on LOAS soft ware, see pecting movements to neighboring territories. www.ecostats.com. Because juveniles and helpers made repeated and extended homesteading movements to Literature Cited sett lement areas before dispersal, prospecting may allow birds to gather information about Baner, B., A. Balt, and E. Diebold, Eds. 1998. localized resources and conspecifi cs before Micronesian Kingfi sher Species Survival 392 Kesler and Haig [Auk, Vol. 124

Plan: Husbandry Manual. Zoological Society Clark, R. G., and D. Shutler. 1999. Avian habi- of Philadelphia, Pennsylvania. tat selection: Patt ern from process in nest- Bensch, S., D. Hasselquist, B. Nielsen, and B. site use by ducks? Ecology 80:272–287. Hansson. 1998. Higher fi tness for philo- Clobert, J., J. O. Wolff, J. D. Nichols, E. patric than for immigrant males in a semi- Danchin, and A. A. Dhondt. 2001. Intro- isolated population of Great Reed Warblers. duction. Pages xvii–xxi in Dispersal (J. Evolution 52:877–883. Clobert, E. Danchin, A. A. Dhondt, and J. D. Bollmann, K., H-U. Reyer, and P. A. Nichols, Eds.). Oxford University Press, Brodmann. 1997. Territory quality and Oxford, United Kingdom. reproductive success: Can Water Pipits Conradt, L., P. A. Zollner, T. J. Roper, K. Anthus spinole a assess the relationship Frank, and C. D. Thomas. 2003. Foray reliably? Ardea 85:83–98. search: An eff ective systematic disper- Boulinier, T., and E. Danchin. 1997. The use of sal strategy in fragmented landscapes. conspecifi c reproductive success for breed- American Naturalist 161:905–915. ing patch selection in terrestrial migratory DeLay, L. S., J. Faaborg, J. Naranao, S. M. P , species. Evolutionary Ecology 11:505–517. T. deVries, and P. G. Parker. 1996. Paternal Bowen, B. S., R. R. Koford, and S. L. Vehrencamp. care in the cooperatively polyandrous 1989. Dispersal in the communally breeding Galapagos Hawk. Condor 98:300–311. Groove-billed Ani (Crotophaga sulcirostris). Dickinson, J. L., and B. J. Hatchwell. 2004. Condor 91:52–64. Fitness consequences of helping. Pages 48– Breininger, D. R., and D. M. Oddy. 2004. Do 66 in Ecology and Evolution of Cooperative habitat potential, population density, and Breeding in Birds (W. D. Koenig and J. L. fi res infl uence scrub-jay source–sink dynam- Dickinson, Eds.). Cambridge University ics? Ecological Applications 14:1079–1089. Press, Cambridge, United Kingdom. Brooker, M., and I. Rowley. 1995. The sig- Doolan, S. P., and D. W. Macdonald. 1996. nifi cance of territory size and quality in the Dispersal and extra-territorial prospecting mating strategy of the Splendid Fairy-wren. by slender-tailed meerkats, Suricata suri- Journal of Animal Ecology 64:614–627. ca a, in the southwestern Kalahari. Journal Brown, J. L. 1964. The evolution of diversity in of Zoology (London) 240:59–73. avian territorial systems. Wilson Bulletin 76: Ekman, J., V. Baglione, S. Eggers, and M. 160–169. Griesser. 2001. Delayed dispersal: Living Brown, J. L. 1974. Alternate routes to sociality under the reign of nepotistic parents. Auk in jays—With a theory for the evolution of 118:1–10. altruism and communal breeding. American Ekman, J., L. Dickinson, J. Hatchwell, and M. Zoologist 14:63–80. Griesser. 2004. Delayed dispersal. Pages 35– Brown, J. L. 1987. Helping and Communal 47 in Ecology and Evolution of Cooperative Breeding in Birds: Ecology and Evolution. Breeding in Birds (W. D. Koenig and J. L. Princeton University Press, Princeton, New Dickinson, Eds.). Cambridge University Jersey. Press, Cambridge, United Kingdom. Brown, J. L., D. D. Dow, E. R. Brown, and S. D. Emlen, J. T., and L. W. Oring. 1977. Ecology, Brown. 1978. Eff ects of helpers on feeding sexual selection, and the evolution of mat- of nestlings in the Grey-Crowned Babbler ing systems. Science 197:215–223. (Pomatostomus temporalis). Behavioral Emlen, S. T. 1982. The evolution of helping. I. Ecology and Sociobiology 4:43–59. An ecological constraints model. American Buden, D. W. 2000. A comparison of 1983 and Naturalist 119:29–39 1994 bird surveys of Pohnpei, Federated Emlen, S. T. 1997. Predicting family dynam- States of Micronesia. Wilson Bulletin 112: ics in social vertebrates. Pages 228–253 in 403–410. Behavioural Ecology: An Evolutionary Byrkedal, I., G. B. Grnstol, T. Lisleand, Approach, 4th ed. (J. R. Krebs and N. B. K. M. Pedersen, H. Sandik, and S. Davies, Eds.). Blackwell Science, Oxford, Stalheim. 1997. Mating systems and terri- United Kingdom. tory in Lapwings Vanellus vanellus. Ibis 139: Engbring, J., F. L. Ramsey, and V. J. Wildman. 129–137. 1990. Micronesian Forest Bird Surveys, the April 2007] Movement and Space Use in Micronesian Kingfi shers 393

Federated States: Pohnpei, Kosrae, Chuuk, Acorn Woodpecker. Proceedings of the and Yap. U.S. Fish and Wildlife Service, National Academy of Sciences USA 99: Honolulu, Hawaii. 7178–7183. Fedy, B. C., and B. J. M. Stutchbury. 2004. Haydock, J., and W. D. Koenig. 2003. Patt erns of Territory switching and fl oating in White- reproductive skew in the polygynandrous bellied Antbird (Myrmeciza longipes), a resi- Acorn Woodpecker. American Naturalist dent tropical passerine in Panama. Auk 121: 162:277–289. 486–496. Hion, M. A. 1980. Food production and com- Forbes, S., R. Grosshans, and B. Glassey. 2002. petitor density as the determinants of feed- Multiple incentives for parental optimism ing territory size. American Naturalist 115: and brood reduction in blackbirds. Ecology 510–530. 83:2529–2541. Hooge, P. N., and B. Eichenlaub. 1997. Animal Fry, C. H., and K. Fry. 1992. Kingfi shers, Bee- Movement Extension to ARCVIEW, version Eaters, and Rollers. Princeton University 1.1. Alaska Science Center, Biological Science Press, Princeton, New Jersey. Offi ce, U.S. Geological Survey, Anchorage, Gaston, A. J. 1978. The evolution of group ter- Alaska. ritorial behavior and cooperative breeding. Johnson, D. M., and C. C. Horit. 2005. American Naturalist 112:1091–1100. Estimating postnatal dispersal: Tracking the Gerlach, G., and S. Bartmann. 2002. unseen dispersers. Ecology 86:1185–1190. Reproductive skew, costs, and benefi ts of Kesler, D. C. 2002. Nest site selection in coop- cooperative breeding in female wood mice eratively breeding Pohnpei Micronesian (Apodemus sylvaticus). Behavioral Ecology Kingfi shers (Halcyon cinnamomina reichen- 13:408–418. bachii): Does nest-site abundance limit Greenwood, P. J. 1980. Mating systems, philop- reproductive opportunities? M.S. thesis, atry and dispersal in birds and mammals. Oregon State University, Corvallis. Animal Behaviour 28:1140–1162. Kesler, D. C., and S. M. Haig. 2004. Thermal Greenwood, P. J., and P. H. Harey. 1982. characteristics of wild and captive The natal and breeding dispersal of birds. Micronesian Kingfi sher nesting habitats. Annual Review of Ecology and Systematics Zoo Biology 23:301–308. 13:1–21. Kesler, D. C., and S. M. Haig. 2005a. Microclimate Griffin, A. S., J. M. Pemberton, P. N. M. and nest-site selection in Micronesian king- Brotherton, G. McIlrath, D. Gaynor, R. fi shers. Pacifi c Science 59:499–508. Kansky, J. O’Riain, and T. H. Clutton- Kesler, D. C., and S. M. Haig. 2005b. Selection Brock. 2003. A genetic analysis of breeding of arboreal termitaria for nesting by coop- success in the cooperative meerkat (Suricata eratively breeding Micronesian Kingfi shers surica a). Behavioral Ecology 14:472–480. Todiramphus cinnamominus reichenbachii. Ibis Haig, S. M., and J. D. Ballou. 1995. Genetic 147:188–196. diversity in two avian species formerly Koenig, W. D., P. N. Hooge, M. T. Stanback, endemic to Guam. Auk 112:445–455. and J. Haydock. 2000. Natal dispersal in the Haig, S. M., J. D. Ballou, and N. J. Casna. 1995. cooperatively breeding Aocorn Wo dpecker. Genetic identifi cation of kin in Micronesian Condor 102:492–502. Kingfi shers. Journal of Heredity 86:423–431. Koenig, W. D., and F. A. Pitelka. 1981. Haig, S. M., J. R. Walters, and J. H. Plissner. Ecological factors and kin selection in the 1994. Genetic evidence for monogamy in evolution of cooperative breeding in birds. the cooperatively breeding Red-cockaded Pages 261–280 in Natural Selection and Woodpecker. Behavioral Ecology and Social Behavior: Recent Research and New Sociobiology 34:295–303. Theory (R. D. Alexander and D. W. Tinkle, Hale, A. M., D. A. Williams, and K. N. Eds.). Chiron Press, New York. Rabenold. 2003. Territoriality and neighbor Koenig, W. D., F. A. Pitelka, W. J. Carmen, assessment in Brown Jays (Cyanocorax morio) R. L. Mumme, and M. T. Stanback. 1992. in Costa Rica. Auk 120:446–456. The evolution of delayed dispersal in Haydock, J., and W. D. Koenig. 2002. cooperative breeders. Quarterly Review of Reproductive skew in the polygynandrous Biology 67:111–150. 394 Kesler and Haig [Auk, Vol. 124

Koenig, W. D., D. Van Vuren, and P. N. Hooge. Mueller-Dombois, D., and F. R. Fosberg. 1998. 1996. Detectability, philopatry, and the distri- Vegetation of the Tropical Pacifi c Islands. bution of dispersal distances in vertebrates. Springer, New York. Trends in Ecology and Evolution 11:514–517. Mulder, R. A., P. O. Dunn, A. Cockburn, Komdeur, J. 1991. Infl uence of territory quality K. A. Laenby-Cohen, and M. J. Howell. and habitat saturation on dispersal options 1994. Helpers liberate female fairy-wrens in the Seychelles Warbler: An experimental from constraints on extra-pair mate choice. test of the habitat saturation hypothesis Proceedings of the Royal Society of London, for cooperative breeding. Pages 1325–1332 Series B 255:223–229. in Acta XX Congressus Internationalis Prt, T. 1994. Male philopatry confers a mat- Ornithologici (B. D. Bell, Ed.). Congress ing advantage in the migratory Collared Trust Board, Wellington, New Zealand. Flycatcher, Ficedula albicollis. Animal Komdeur, J. 1992. Importance of habitat satura- Behaviour 48:401–409. tion and territory quality for evolution of Pasinelli, G., K. Schiegg, and J. R. Walters. cooperative breeding in Seychelles Warbler. 2004. Genetic and environmental infl uences Nature 358:493–494. on natal dispersal distance in a resident bird Komdeur, J. 1994. Experimental evidence for species. American Naturalist 164:660–669. helping and hindering by previous off spring Pitcher, T. E., and B. J. M. Stutchbury. 2000. in the cooperative-breeding Seychelles Extraterritorial forays and male parental Warbler Acrocephalus sechellensis. Behavioral care in Hooded Warblers. Animal Behaviour Ecology and Sociobiology 34:175–186. 59:1261–1269. Langen, T. A., and S. L. Vehrencamp. 1998. Porter, J. H., and J. L. Dooley, Jr. 1993. Animal Ecological factors aff ecting group and ter- dispersal patt erns: A reassessment of ritory size in White-throated Magpie-Jays. simple mathematical models. Ecology 74: Auk 115:327–339. 2436–2443. Li, S.-H., and J. L. Brown. 2000. High frequency Pratt, H. D., P. L. Bruner, and D. G. Berrett. of extrapair fertilization in a plural breeding 1987. A Field Guide to the Birds of bird, the Mexican Jay, revealed by DNA mic- Hawaii and the Tropical Pacifi c. Princeton rosatellites. Animal Behaviour 60:867–877. University Press, Princeton, New Jersey. Ligon, J. D. 1999. The Evolution of Avian Pribil, S., and W. A. Searcy. 2001. Experimental Breeding Systems. Oxford University Press, confi rmation of the polygyny thresh- Oxford, United Kingdom. old model for Red-winged Blackbirds. Ligon, J. D., and S. H. Ligon. 1990. Green Proceedings of the Royal Society of London, Woodhoopoes: Life history traits and soci- Series B 268:1643–1646. ality. Pages 33–65 in Cooperative Breeding Pryke, S. R., and S. Andersson. 2003. Carotenoid- in Birds: Long-term Studies of Ecology based epaulett es reveal male competitive and Behavior (P. B. Stacey and W. D. ability: Experiments with resident and Koenig, Eds.). Cambridge University Press, fl oater Red-shouldered Widowbirds. Animal Cambridge, United Kingdom. Behaviour 66:217–224. Lima, S. L. 1984. Territoriality in variable Rappole, J. H., and A. R. Tipton. 1991. New har- environments: A simple model. American ness design for att achment of radio trans- Naturalist 124:641–655. mitt ers to small passerines. Journal of Field Lhmus, A., and Ü. Vli. 2004. The eff ects of Ornithology 62:335–337. habitat quality and female size on the pro- Reed, J. M., T. Boulinier, E. Danchin, and ductivity of the Lesser Spott ed Eagle Aquila L. W. Oring. 1999. Informed dispersal: pomarina in the light of the alternative prey Prospecting by birds for breeding sites. hypothesis. Journal of Avian Biology 35: Pages 189–259 in Current Ornithology, vol. 455–464. 15 (V. Nolan, Jr., E. D. Kett erson, and C. F. Luck, G. W. 2003. Diff erences in the reproduc- Thompson, Eds.). Plenum Press, New York. tive success and survival of the Rufous Richardson, D. S., F. L. Jury, K. Blaakmeer, J. Treecreeper (Climacteris rufa) between a Komdeur, and T. Burke. 2001. Parentage fragmented and unfragmented landscape. assignment and extra-group paternity in a Biological Conservation 109:1–14. cooperative breeder: The Seychelles Warbler April 2007] Movement and Space Use in Micronesian Kingfi shers 395

(Acrocephalus sechellensis). Molecular Ecology information about life history parameters? 10:2263–2273. Ecological Applications 11:1722–1737. Roshier, D. A., and J. R. W. Reid. 2003. On ani- U.S. Fish and Wildlife Service. 1984. mal distributions in dynamic landscapes. Endangered and threatened wildlife and Ecography 26:539–544. plants: Determination of endangered sta- Savidge, J. A. 1987. Extinction of an island forest tus for seven birds and two bats on Guam avifauna by an introduced snake. Ecology and the Northern Mariana Islands. Federal 68:660–668. Register 50 CFR Part 17 49:33881–33885. Schjørring, S., J. Gregersen, and T. Bregnballe. U.S. Fish and Wildlife Service. 2004. Draft 1999. Prospecting enhances breeding suc- Revised Recovery Plan for the Sihek or cess of fi rst-time breeders in the Great Guam Micronesian Kingfi sher (Halcyon Cormorant, Phalacrocorax carbo sinensis. cinnamomina cinnamomina). U.S. Fish and Animal Behaviour 57:647–654. Wildlife Service, Portland, Oregon. [Online.] Schoener, T. W. 1983. Simple models of opti- Available at www.epa.gov/fedrgstr/EPA- mal feeding-territory size: A reconciliation. SPECIES/2004/April/Day-28/e9585.htm. American Naturalist 121:608–629. Vehrencamp, S. L. 1980. To skew or not to skew? Selander, R. K. 1964. Speciation in Wrens of Pages 869–874 in Acta XVII Congressus the Genus Campylorhynchus. University Internationalis Ornithologici (R. Nöhring, of California Publications in Zoology 74: Ed.). Deutsche Ornithologen-Gesellschaft , 1–305. Berlin. Skutch, A. F. 1935. Helpers at the nest. Auk 52: Walters, J. R. 2000. Dispersal behavior: An orni- 257–273. thological frontier. Condor 102:479–481. Smith, R. D., and N. B. Metcalfe. 1997. Where Walters, J. R., C. B. Cooper, S. J. Daniels, G. and when to feed: Sex and experience aff ect Pasinelli, and K. Schiegg. 2004. Conservation access to food in wintering Snow Buntings. biology. Pages 197–227 in Ecology and Behaviour 134:143–160. Evolution of Cooperative Breeding in Birds Stacey, P. B. 1979. Habitat saturation and com- (W. D. Koenig and J. L. Dickinson, Eds.). munal breeding in the Acorn Woodpecker. Cambridge University Press, Cambridge, Animal Behaviour 27:1153–1166. United Kingdom. Stacey, P. B., and W. D. Koenig, Eds. 1990. Walters, J. R., C. K. Copeyon, and J. H. Carter Cooperative Breeding in Birds: Long- III. 1992. Test of the ecological basis of Term Studies of Ecology and Behavior. cooperative breeding in Red-cockaded Cambridge University Press, Cambridge, Woodpeckers. Auk 109:90–97. United Kingdom. White, G. C., and R. A. Garrott. 1990. Analysis Stacey, P. S., and J. D. Ligon. 1991. The benefi ts- of Wildlife Radio-tracking Data. Academic of-philopatry hypothesis for the evolution of Press, San Diego, California. cooperative breeding: Variation in territory Zack, S. W. 1990. Coupling delayed breeding quality and group size eff ects. American with short-distance dispersal in coopera- Naturalist 137:831–846. tively breeding birds. Ethology 86:265–289. Stamps, J. A., and V. V. Krishnan. 1999. A learn- Zwicker, S. M., and J. R. Walters. 1999. ing-based model of territory establishment. Selection of pines for foraging by Red- Quarterly Review of Biology 74:291–318. cockaded Woodpeckers. Journal of Wildlife Tyre, A. J., H. P. Possingham, and D. B. Management 63:843–852. Lindenmayer. 2001. Inferring process from patt ern: Can territory occupancy provide Associate Editor: M. Bri ingham