QL 671 ,W7 v. 124 no.l he Wilson Journal of Ornithology Volume 124 , Number 1 , March 2012 £lvsH Sato Stewart Library KxK l'? %j3 Academy of Natural Science* of Philadelphia Published by the Wilson Ornithological Society THE WILSON ORNITHOLOGICAL SOCIETY FOUNDED 3 DECEMBER 1888 Named after ALEXANDER WILSON, the first American ornithologist. ¥\J0 • I President — Robert C. Beason. P. O. Box 737, Sandusky, OH 44871, USA: e-mail: Robert. C.Beason@ gmail.com First Vice-President — Robert L. Curry, Department of Biology, Villanova University, 800 Lancaster Avenue, Villanova, PA 19085, USA; e-mail: robert.curry@villanova.edu Second Vice-President — Sara R. Morris, Department of Biology, Canisius College, 2001 Main Street, Buffalo, NY 14208, USA, e-mail: morris@canisius.edu Editor — -Ciait E. Braun, 5572 North Ventana Vista Road, Tucson. AZ 85750. USA; e-mail: TWILSONJO@ comcast.net Secretary — John A. Smallwood, Department of Biology and Molecular Biology, Montclair State University, Montclair. NJ 07043, USA; e-mail: smallwoodj@montclair.edu Treasurer — MelindaM. Clark, 52684 Highland Drive, South Bend, I N46635, USA; e-mail: MCIark@tcservices.biz Elected Council Members — Mary Bomberger Brown, Mary C. Garvin, Timothy J. O’Connell, and Mark S. Woodrey (terms expire 2012); Mark E. Dcutschlander and Rebecca J. Safran (terms expire 2013); William H. Barnard, Margret I. Hatch, and Mark E. Hauber (terms expire 2014). Living Past-Presidents — Pershing B. HofsJund, Douglas A. James, Jerome A. Jackson, Ciait E. Braun, Richard C. Banks, Richard N. Conner. Keith L. Bildstein, Edward H. Burtt Jr., John C. Kricher. William E. Davis Jr., Charles R. Blem, Doris J. Watt, James D. Rising, and E. Dale Kennedy. Membership dues per calendar year are: Regular. $40.00: Student, $20.00; Family, $50.00; Sustaining, $100.00; Life memberships. $1,000.00 (payable in four installments). The Wilson Journal of Ornithology is sent to all members not in arrears for dues. THE WILSON JOURNAL OF ORNITHOLOGY (formerly The Wilson Bulletin) THE WILSON JOURNAL OF ORNITHOLOGY (ISSN 1559-4491) is published quarterly in March, June, September, and December by the Wilson Ornithological Society, 810 East 10th Street. Lawrence. KS 66044-8897. The subscription price, both in the United States and elsewhere, is $40.00 per year. Periodicals postage paid at Lawrence. KS. POSTMASTER: Send address changes to OSNA. 5400 Bosque Boulevard. Suite 680. Waco. TX 76710. All articles and communications for publication should be addressed to the Editor. Exchanges should be addressed to The Josselyn Van Tyne Memorial Library, Museum of Zoology. Ann Arbor, MI 48109, USA. Subscriptions, changes of address, and claims for undelivered copies should be sent to OSNA, 5400 Bosque Boulevard. Suite 680, Waco. TX 76710. USA. Phone: (254) 399-9636: e-mail: business@osnabirds.org. Back issues or single copies are available for $ 1 2.00 each . Most back issues of the journal are available and may be ordered from OSNA. Special prices will be quoted for quantity orders All issues of the journal published before 2000 are accessible on a free web site at the University of New Mexico library (http://elibrary.unni.edu/sora/). The site is fully searchable, and full-text reproductions of all papers (including illustrations) arc available as either PDF or Dj Vu files. © Copyright 2012 by the Wilson Ornithological Society Printed by Allen Press Inc., Lawrence, KS 66044, USA. LR. Wilson s Snipe ( Gallinago delicata). Illustration by Scott Rashid. ® Th'S P°Per me<”S ,he O' ANSI/N.SO Z39.48-I992 (Permanence o, Paper) tj 1 watcr^ns^mV^- J'he Northern Blac^ Swi,t (O pseloides niger borealis) is known to nest in moist and cool sites near placed on Northern^ B hok^ r^ °! win.tering areas and migra‘»on paths is non-existent. Use of light-level geolocators revealed the first known V- \W* . borealis in Costa Rica in spring had no specimens collected for confirmation (Stiles and Skutch 1990). Kiff (1975) tentatively assigned a female swift col¬ lected in Costa Rica to C. n. borealis based on 1 2 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 wing and tail measurements that are within ranges for C. n. costarieensis, further highlighting the uncertainty of migration and winter distribution for this subspecies. Winter records for Northern Black Swift arc non-existent. Several factors contribute to the lack of knowledge about migration and winter distribu¬ tion of this species, including difficulty in accurate field identification of individuals due to high and rapid flight, problems differentiating this species from similar-size members of Cypseloides that occupy Central and South America, and inability to verify observation records. No band recoveries exist outside of the United States from ~200 Northern Black Swifts banded from 1950 to present (Bird Banding Laboratory, pers. comm.), Currently, satellite tracking devices which provide accurate tracking of individuals are not sufficiently small to place on a species the size of Black Swifts. However, light-level geolocators, devices that record ambient light levels at fixed intervals, are highly effective instruments for tracking long-distance migratory species and are sufficiently small to place on swifts. They are battery-powered instruments with a microproces¬ sor, clock, and memory for data storage; geo¬ graphical positions can be calculated from the data collected by the devices. Geolocators must be retrieved to download data, and the Black Swift is particularly suited to recapture due to its high breeding colony fidelity and an individual propensity to reuse the same nest from year to year (Foerster 1987. Collins and Foerster 1995, Marin 1997, Hirschman et al. 2007). We placed geolocators on four Northern Black Swifts with the objective to gather infor¬ mation about the migratory path, timing, and winter destination of this species. Identifying the connectivity of a migrating species between breeding sites and wintering areas is crucial to understanding the species' ecology and in guiding conservation efforts. Time spent in widely separated and ecologically disparate locations by migrating species and the strength of this link can have great biological consequences for individuals and populations, me uding reproductive success, population dy¬ namics, behavioral ecology, evolution, and re- ^cctive pressures (Webster can nr' . ^Advances in geolocator technology ua"s a?h h rfol™‘ion by individ betaus e 'nf Ut,0n- Wc ‘-™*'ded this study because tnforntatton on migration and wintering areas of Northern Black Swifts was virtually non¬ existent (Lowther and Collins 2002, Wiggins 2004). The small sample does not tell us how- weak or strong the migratory connectivity is for this subspecies, but forms a foundation for additional knowledge of this species' ecology and can guide future studies. METHODS Study Sites. — Northern Black Swifts nest at or near waterfalls typically inaccessible due to steep and vertical configuration (Knorr 1961. Levad et al. 2008). More than 100 breeding sites of this species have been documented in North America (Lowther and Collins 2002, Levad et al. 2008) with only a few records of alternate types of sites such as sea caves in California (Legg 1956). small cave-like boulder configurations in streams (Foer¬ ster and Collins 1990, Johnson 1990, Hurtado 2002), and caves (Davis 1964, Northern British Columbia Caving Club 2003). We chose Fulton Resurgence Cave (39 49' N. 107 24' W) and Ouray Box Canyon Falls (38 I' N, 107 40' W) in Colorado because of accessi¬ bility and the probability of capturing and recapturing Black Swifts using hand-held or mist nets. These breeding colonies are two of the largest in Colorado (Levad et al. 2008) with an average of eight nesting pairs (range 7-9, n = 6 yrs) at Fulton Resurgence Cave (KMP, pers. obs.) and an average of 1 1 nesting pairs (range 7- 15, n = 10 yrs) at Box Canyon Falls (Hirshman et al. 2007, Levad et al. 2008). Fulton Resurgence Cave is a limestone cave with a small stream issuing from it. forming a microhabitat conducive for a Black Swift breeding colony (Knorr 1961). Mist nets placed near the mouth of the cave, the only ingress/egress for swifts, have resulted in a recapture rate of 41% since banding of adults began in 2006 (KMP. pers. obs. ). Box Canyon Falls is a popular tourist site and walkways provide views of the falls. The walkways allow access to several nest sites and. with the aid of ladders, several nests can be reached with hand-held nets. Data Collection. — We used four MklOS model geolocators, manufactured by the British Antarc¬ tic Survey (BAS), programmed to continuously measure light levels every minute and archive the maximum measurement for each 10-min period. The devices weighed 1.2 g, measured IS X 9 X 6 mm, and have a light sensor mounted at the tip of a 10-mm stalk angled at 15 to prevent it from being covered by feathers. The instruments are Beason et al. • BLACK SWIFT MIGRATION AND WINTERING AREAS 3 encapsulated in a water-resistant housing with two external terminals for commands and data transfers. We designed a backpack harness system modified from Buehler et al. (1995) because Black Swift legs are too attenuated for the leg- loop harness often used for geolocators on passerines (Rappole and Tipton 1991). The harness material. 5 mm tubular Teflon ribbon (Bally Ribbon Mills. Bally. PA, USA), was attached at four points to the geolocator and crossed under the keel. We secured the free ends of the ribbon with size 69 bonded right twist Kevlar thread (The Thread Exchange Inc., Wea- verville. NC. USA) and stitched the ribbon where ii crossed the keel to avoid shifting. We applied cyanoacrylate glue on all stitches and cut ends to prevent fraying. We attached geolocators to four adult Black Swifts in August 2009. three al Fulton Resurgence Cave (2 females. I male) and one at Box Canyon Falls (male); the birds weighed 49.5 51.5 g. Each geolocator, including harness materials, weighed 1.5 g. representing 2.9-3% of body weight, well within guidelines suggested by Caccamise and Hedin (1985). Delta Analysis. — We conducted pre-deployment calibration for ~9 days and post-deployment calibration for ~7 days by placing them at a known location with a dear view' of the sky. We retrieved Ihree of Ihe four geolocators in July and August 2010. We used software programs (BASTrak) devel¬ oped by BAS to download, process, and interpret data archived by the loggers, each of which had collected data throughout their entire deployment. We rejected latitude data gathered ~30 days before and after the equinoxes since day lengths at the equinoxes are equal at all latitudes, resulting in poor location fixes. Internal clocks maintained accuracy during deployment and there was no need to correct for clock drift. Two values are required for analyzing and plotting the geolocator data: the dusk/dawn light transition threshold and the corresponding sun elevation angle at this threshold. We chose a sensitive light transition threshold value of two to reduce variation in day length due to the effects of shading which influences the resulting distribution of location fixes. We used static pre-deployment calibration to calculate the corresponding sun elevation angles (-6.4 , -6.5 , and -6.6 ). We calculated times of sunrise and sunset using TransEdit2; positions were calculated with Bird- Tracker which derives longitude from absolute time of local midday/midnight and calculates latitude by comparing day/night length, a tech¬ nique which provides two geographical positions/ day. Wc used only midnight fixes to produce maps based on the assumption that sw'ifts were roosting at night and migrated during die day. We identified days with irregular shading events, resulting in shorter day lengths or anomalous transition limes, by visually inspecting sunrise and sunset times and excluded them from the analysis. An average of 145 fixes for each bird remained to map wintering range and an average of 26 fixes remained to map the spring migration path. Mapping of fall migration was not possible due to overlap with the fall equinox. We calculated kernel density surfaces using the wintering area data from each geolocator with the Spatial Analyst Kernel Density function (ESRI 2009). This function calculates density of fixes in a search radius around those fixes. These densities lit a smoothly curved surface over each location. The surface value was highest at the location of the point and diminished with increasing distance from the point. We used a fixed kernel with a search radius of 185 km to compensate for the approximate average error in latitude and longi¬ tude known to occur in geolocator data (Phillips et al. 2004). The kernel function is based on the quadratic kernel function described in Silverman (1986). We calculated the density surfaces at 1-km resolution as this is adequate to capture density at a small scale over a large geographic area. We calculated 90%. 75%. and 50% density polygons from the kernel density surfaces to enhance graphic displays ot higher use density areas. We used the average nearest-neighbor distance function in Arclnfo Spatial Statistics (ESRI 2009) to characterize the spatial point pattern of winter locations. This function quanti¬ fies and characterizes the spatial pattern of each geolocator and indicates if the pattern is evenly dispersed, random, or clustered compared to a spatial random distribution. We estimated approx¬ imate migration duration, arrival, and departure events from plotting longitude and date. We used the 50% kernel density polygons for all three swifts to describe land cover use and overlaid those with a global land cover layer usine 2009 satellite imagery at a 300-m resolution produced by the European Space Agency Glob- Cover 2009 Project (Bontemps et al. 2010). 4 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124, No. 1. March 2012 TABLE 1 . Phenology of migration stages of three North¬ ern Black Swifts". Departure from Colorado Arrival at wintering area Time spent at wintering area Departure from wintering area Arrival in Colorado Duration of southbound migration Duration of northbound migration 14 Sep (10-19 Sep) 5 Oct (28 Sep-12 Oct) 220 days (209 to 227 days) 13 May (9-20 May) 1 Jun (23 May-18 Jun) 2 1 days ( 1 8 to 23 days) 20 days (14 to 29 days) “ Arrival and departure dates are presented as means with range in parenthesis. RESULTS Geolocators recovered from two females at Fulton Resurgence Cave and one male at Box Canyon Falls represent a 15% recovery rate. The Black Swifts initiated fall migration from Colo¬ rado beginning on 10 September and continued through 19 September 2009. We used longitudinal information around the time of the autumnal equinox and documented the swifts arrived at their wintering location in South America be¬ tween 28 September and 12 October 2009. Approximate dates of migration initiation north from wintering areas began on 9 May and continued through 20 May 2010. Dates of arrival at breeding sites began 23 May and continued through 18 June 2010 (Table L Fig. I). Kernel density estimates indicate all three birds over-wintered primarily in the lowland rainforest of western Brazil with some kernels extending into Bolivia, Colombia. Peru, and Venezuela (Fig. 2). Average nearest-neighbor distance anal¬ ysis for all geolocators exhibited clustering with nearest-neighbor ratios = 0.89 ( P = 0.02), 0.74 ( P < 0.001), and 0.77 (P < 0.001) for geolocator #553, #554. and #556, respectively. The distance between the Ouray Box Canyon Falls breeding site and the center of the wintering range in Brazil (#554) is 6,901 km and the average distance between the Fulton Resurgence Cave breeding site and the center of the wintering range in Brazil (#553 and #556) is 7.025 km. The swifts traveled at an average speed of 341 km/day during the 2009 fall migration and an average speed of 393 km/day duiing the 2010 spring migration. The inaccuracy ol geolocators precludes precise calculation of an average daily distance covered by each bird. The land cover overlay maps for 50% kernel density areas for all three birds indicate a FIG. 1 . Spring migration routes for three (#'s 553. 554. 5561 individual Northern Black Swifts marked in Colorado. dominant land cover (>86%) of closed to open broadleaved evergreen or semi -deciduous forest and a small percentage (2-10%) of the kernel density areas are classified as closed to open Beason et al. • BLACK SWIFT MIGRATION AND WINTERING AREAS 5 isws SOUTW 7*WW FIG. 2. Northern Black Swifts marked in Colorado wintering distribution kernel density contours (red = 50%. green = 15%. blue = 90%) from 28 September- 1 2 October 2009 to 9-20 .May 2010. broadleaved forest regularly flooded. Areas of mosaic cropland/vegetation, mosaic forest shrub- land/grassland, closed to open shrubland or grass¬ land. bare areas, and water bodies represented <2% use. DISCUSSION We documented the timing of fall and spring migration, wintering area, and spring migration paths for three Northern Black Swifts using geolocators. The return dates to breeding sites after spring migration and dates of the initiation of fall migration correlate with data collected in other research (Hirschman et al. 2007). Wintering area locations were previously unknown, and only sporadic reports existed which did not fully delineate migratory paths. The highest kernel density estimates indicate the three birds wintered almost entirely within the State of Amazonas, Brazil. The clustering of the three individuals exhibited by nearest-neighbor analysis might be expected based on the birds seeking a physiologically optimal climate, pre¬ ferred habitat, abundant prey, or other factors within a certain geographic area. Amazonas is composed almost entirely (—98%) of lowland rainforest at elevations between 34 and 116 m. The area is sparsely populated with a density of 2.05 inhabitants/km2 with 78% of die population in cities (IBGE 201 1). Amazonas has an equato¬ rial tropical rainforest climate with annual rainfall of 1.50-2.50 m and all months have a mean precipitation of at least 60 mm (IBGE 201 1 ). The average temperature per day per year is 26.7 C (23.3-31.4 Cl with high humidity (Brasil Travel Guide 2011). The three birds tracked in this study represent only a small geographical subset of the Northern Black Swift population and further studies are needed to delineate more completely the full extent of the wintering distribution of this subspecies. The three birds wintered in the same general area, suggesting a high level of connec¬ tivity between breeding and wintering popula¬ tions. Stutchbury et al. (2009) found a similar connectivity for Wood Thrush ( Hylocichla mus- lelina), not previously documented for migratory songbirds. The large w intering areas may retied a temporal movement noted for each bird.' The data indicate a trend for each bird to be in the eastern portion of the kernels in October with gradual movement west in April and May. The most likely explanation for this replicated non-random change 6 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124, No. 1, March 2012 in position is that the birds moved, perhaps following their food source. This net westward directional movement is less likely to be an artifact due to shading or weather-related varia¬ tion because it was consistent among the three birds tracked. Northern Black Swift roost sites and roosting behavior in South American wintering areas are unknown. Waterfalls, caves, and dripping rock faces serve as roosting and breeding sites in breeding areas. There is only one documented observation of a roost site for Northern Black Swifts in South America, discovered during fall 1992 and 1993, on the walls of a steep gorge along the Rio Caucu in the foothills of Colombia. Black Swifts roosted consecutively at dusk for a week in a compact group, mainly with White- collared Swifts ( Slreptoproche zonaris ), clinging to the volcanic rock of a 40- m cliff overlooking the river, indicating that rocky river banks are used as roosting sites during migration (Stiles and Negret 1994). Similar .sites may be used in wintering areas if available, but this information is completely lacking. Non-brccding Common Swifts (Apus apus) are known to ‘roost* aerially in breeding areas and it is believed they spend ~9 months of the year continuously on the wing. Non-breeding birds may fly continuously for several years (Backman and Alerstam 2001, Tarburton and Kaiser 2001), Common Swifts are also known to occasionally roost at night by hanging on the foliage of trees ( Holmgren 2004). Any of these scenarios is possible for Northern Black Swifts in wintering areas. Foraging activities of the Northern Black Swift in wintering areas are unknown but kernel densities indicate the swifts range over a large area in winter, suggesting the birds spend a lot of time on the wing. The capability ot geolocators for tracking small birds is still being explored and the potential is great. However, the devices are not without limitations. A major obstacle for success is that once the devices are deployed, the bird must be recaptured after a complete migration cycle has occurred to obtain data. Our study indicates the suitability of the Northern Black Swift for geolocator deployment and recapture, primarily due to this species* strong nesting colony fidelity and abihty to carry small devices for lone periods Calculation of latitude is unreliable around equmox and near the equator because lengt day and mght is equal. The accuracy of calcul day length is especially affected for terrestrial species hy shading factors that alter recorded light levels such as cloudy weather, foliage, and topographic shading of roost sites, resulting in latitude uncertainties. Fudickar el al. (2011) found the devices had an error of 201 ± 43 km for latitude and 12 ± 3 km for longitude (±95% Cl) for stationary geolocators (n = 30) in forested habitat. The apparent retreat of bird #554 from Colorado to the Pacific Ocean south of Baja California (Fig. 1) during spring migration is the result of one data point and the accuracy of this fix is questionable. It may or may not represent an actual movement by the bird and is possibly the result of an extended period of shading. We did not eliminate this position fix since the total day length did not drastically differ from the other day lengths of that time period. Accurate longitudinal information can be ascer¬ tained as this is not affected by equinox and we successfully used longitude near the autumnal equinox to indicate when the birds arrived at their wintering location. Black Swift breeding require¬ ments, such as nesting behind waterfalls in deeply shaded niches in steep and narrow canyons, or in caves where the performance of geolocators is often compromised by darkness, resulted in some unusable data during the breeding season. Docu¬ mented Black Swift nocturnal roosting behavior during migration is limited and indicates this could be a factor influencing the effectiveness of geolocators for tracking this species. If winter nocturnal roost sites are similar to those docu¬ mented in migration and al breeding sires, this will also influence the accuracy of the data collected by geolocators. Despite these limitations, geolocators far surpass hand recovery information or depen¬ dence on sporadic sightings to identify migratory paths and winter distribution of the Northern Black Swift. Understanding the theory behind geolocation is extremely important for interpreting and using the data collected to produce maps showing animal movements (Hill 1994). Once the theory is understood, knowledge of the behavior of the animal being studied and of weather patterns in the area where the animal was tracked can be used to provide insight into movement patterns. The mapped winter range of Black Swifts is an area that typically experiences high cloud cover. Thus, a significant number of the location fixes are most likely shifted to the north artificially because of cloud cover in the winter range as compared to the Season et at. • BLACK SWIFT MIGRATION AND WINTERING AREAS 7 Colorado calibration location. Therefore, the southern portion of the mapped winter range is most likely the area where the swifts spent the winter. The technical limitations of geolocalors and lack of knowledge of Black Swift behavior in wintering areas, such as daily foraging (light distance, and roosting locations and tinting further confound data interpretation. The Black Swift is protected under the Migratory Bird Treaty Act in the United States and the Convention for the Protection of Migra¬ tory Birds and Game Mammals in Mexico. This study documents Northern Black Swifts spending -'220 days in Brazil during winter 2009-2010. the first records of the species in this country. This study identifies an annual non-breeding geogra¬ phic area of the Northern Black Swift and is a significant step toward conservation of this species. Future studies could include use of geolocators on subsets of Northern Black Swifts from other areas of North America which would help delineate the strength of migration connectivity for this subspecies. Development of satellite transmitters small enough for use on Black Swifts will provide greater accuracy than geolocalors and can possibly answer questions about roosting and foraging behavior. CONSERVATION IMPLICATIONS Knowledge of migratory pathways and winter distribution of a species enables evaluation of those geographical areas, including ecologic analysis and research, identification of potential habitat threats, and development of conservation strategies. The homogeneity of the wintering areas for Northern Black Swifts evidenced in this study suggests limited winter resource use by this subspecies, which could have long-term conser¬ vation impacts. The current rate of deforestation in Brazil could directly threaten this subspecies. One of the most refined computer models for simulating deforestation. SimAmazonia I. indi¬ cates the rale of deforestation in the State of Amazonas will increase rapidly in the coming decades which could result in a loss of up to 30% of the forest cover by 2050 (Soares-Filho et al. -006). Climate change and global warming predictions also pose threats to habitat and prey availability for this subspecies. Roberson and Collins (2008) identified declines in some North¬ ern Black Swift populations but it is unknown if declines are due to environmental problems in breeding areas, during migration, in wintering areas, or some combination of these possibilities. ACKNOWLEDGMENTS The authors thank Black Canyon Audubon Society, Evergreen Audubon Society. Grand Valley Audubon Society. Roaring Fork Audubon Society. Colorado Field Ornithologists, and the Colorado Chapter of The Wildlife Society for contributions to this project. D. M. Elwonger, II. E. Kingery. L. R Patrick, A. R. Robinsong. and W. P. Schmoker contributed funds to the Riehaal G. Levad Memorial Fund held in trust at the Rocky Mountain Bird Observatory to be used specifically for this project. The Grand River Hospital perioperative staff contributed funds to the M. A. Potter Memorial Fund to be used specifically for this project. Significant assistance with interpretation of geolocator data was received from N. B. F. Cousens and D. M. Morrison. A. O. Panjabi and L. L. Jenks provided thoughtful reviews of the manuscript. N. M. Goedert. T. W. Patrick, and C. W. Reichert contributed invaluable support during Black Swift banding expeditions to Fulton Resur¬ gence Cave. We thank die White River National Forest for project support and logistics. K. It. Knudsen for access to research library resources, and S. E. Hirshnian for coordinating access to Box Canyon Falls. Knowledgeable suggestions by peer reviewers C. T. Collins and A. M. Fudickar were invaluable. This publication is dedicated to the memory and spiritual guidance of our friend and colleague Richard G. Levad. LITERATURE CITED Backman. .1. AND T. AlERSTAM. 2001. Confronting the winds: orientation and flight behavior of roosting Swifts, A pus a pus. Proceedings of the Royal Society of London, Series B 268:1081-1087. Bontkmps, S.. P. Dp.fourny. and F. Van Boc.aert. 2010. GLOBCOVFR 2009. Version GLOBCOVER2(X)9_ PDM_1.0. European Space Agency and University Catholique de Louvain. Belgium. www.esa.int/esaCP/ SFM5N3TRJ HGJndcx_ 1 .html Brasil Travel Guide. 2011. Climate, temperatures and rain in Brazil. Brasil Travel Guide, Rio de Janeiro. www.brazil-travel-guide.eom/Brazil-Map-Weather. html# Amazon Brazilian Institute of Geography and Statistics (IBGE). 201 1 Portal. Brazil, www.ibgc.gov.br/english/ Buchanan. O. M. and H. L. Fierstine. I%4. Another Pacific record of the Black Swill off Mexico. Condor 64:161-162. Buehler. D. a.. J. D. Fraser. M. R. Fuller, L. S. McAllister, and J. K. D. Seegar. 1995. Captive and field-tested radio transmitter attachment for Bald Eagles. Journal of Field Ornithology 66:173-180. Caccamise. D. F. AND R. s. Hed/n. 1985. An aerodynamic basis tor selecting transmitter loads in birds Wilson Bulletin 97:306-318. Collins, C. T. and K. S. FoERsrm. 1995. Nest site fidelity and adult longevity in the Black Swift (Cypseloides niger). North American Bird Bander 20:11-14 8 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 Davidson, M. E. M. 1934. An occurrence of the Northern Black Swift off the Guatemalan Coast. Condor 36:37. Davis. D. G. 1964. Black Swifts nesting in a limestone cave in Colorado. Wilson Bulletin 76:295-296. Environmental Systems Research Institute (ESRI). 2009. Arclnfo Desktop. Version 9.3.1. Environmental Systems Research Institute. Redlands, California. USA. Foerster, K. S. 1987. The distribution and breeding biology of the Black Swift (Cypselnides niger) in southern California. Thesis. California State University, Long Beach. USA. Foerster, K S. and C. T. Collins. 1990. Breeding distribution of the Black Swift in southern California. Western Birds 21:1-9. Fudickar, A. M.. M. Wikelski. and J. Partecke. 2011. Tracking migratory songbirds: accuracy of light-level loggers (geolocators) in forest habitats. Methods in Ecology and Evolution: DOI: 10.1 1 I I /j.204 1 2 1 OX. 201 1.001 36x. Hill, R. D. 1994. Theory of geolocation by light levels. Pages 227-236 in Elephant seals: population ecology, behavior and physiology (B. .1. Le Boeuf and R. M, Laws, Editors). University of California Press. Berkeley, USA. Hirshman. S. E., C. Gunn, and R. G. Levad. 2007. Breeding phenology and success of Black Swifts in Box Canyon. Ouray, Colorado. Wilson Journal of Ornithology 119:678-685. Holmgren, J. 2004. Roosting in tree foliage by Common Swifts Apus apus. Ibis 146:404-416. Hurtado, P. 2002. Probable Black Swift ( Cypselnides niger) nesting colony found in the Wet Mountains. Pueblo County. Journal of Colorado Field Ornitholo¬ gists 36:60-61. Johnson. P. W 1990. Black Swift ( Cypselokles niger) nesting in the Jemez Mountains of New Mexico. New Mexico Ornithological Society Bulletin 18:13-15. Kief. L. F. 1975. Notes on southwestern Costa Rican birds. Condor 77:101-103. Knorr. O A. 1961, The geographical and ecological distribution of the Black Swift in Colorado. Wilson Bulletin 73:155-170, Legg, K. 1956. A sea-cave nest of the Black Swift. Condor 58:183-187. Levad. R. G . K. M. Potter, C. W. Shultz. C. Gunn, and J. G. Doerr. 2008. Distribution, abundance, and nest- site characteristics of Black Swifts in the southern Rocky Mountains of Colorado and New Mexico. Wilson Journal of Ornithology 120:331-338. Lowther. P. e. and C. T. Collins. 2002. Black Swift (Cypselnides niger). The birds of North America. Number 676. Marin, M. 1997. Some aspects of the breeding biology of the Black Swift. Wilson Bulletin 109:290-306. Northern British Columbia Caving Club. 2003. The Redemption Cave Project. Prince George. B.C.. Canada. www.cancaver.ca/sciences/Redempl/REDWEBPAGEI. htm Phillips, R. A.. J. R. D. Silk. J. P. Ckoxall. v, Afanasyev, wn D. R. Briggs. 2004. Accuracy nl geolocation estimates for (lying seahirds. Marine Ecology Progress Series 266:265-272. Rappole, J H. and A. R. Tipton. 1991. New hames* design for attachment of radio transmitters to small passerines. Journal of Field Ornithology 62:335-337. Roberson, D. and C. T Coi lins. 2008. Black Swift ( Cypselnides niger). Pages 245-249 in California bird species of special concern: a ranked assessment of species, subspecies, and distinct populations of binF of immediate conservation concern in California. Studies of Western Birds Number I (W. D. Shulnrd and T. Gardali. Editors). Western Field Ornithologists. Camarillo. California and California Department ol Fish and Game, Sacramento, USA. SILVERMAN. B W. 1986. Density estimation for statistics and data analysis. Chapman and Hall. New Vork. USA. Soares- Filho. B. S„ D. C. Nepstad, L. M. Curran. G. C. Cerqueira, r. a. Garcia, C. A. Ramos, E. Voll, A. McDonald, P. Li-pebvre. and P. Schlesinger. 2006. Modeling conservation in the Amazon Basin. Nature 440:520-523. Stiles. F. G. and A. .1. Negret. 1994. The nonbreeding distribution of the Black Swift: a clue from Colombia and unsolved problems. Condor 96:1091-1094. St ii i s, F. G. and A. F. SklTTCH. 1990. A guide to the birds of Costa Rica. Cornell University Press. Ithaca. New York, USA. Stuichbury. B. J. M„ S. A. Tarof, T. Done, E. Gov\. P M. Kramer, J. Taltin, J. W. Fox, and V. Afanasyev. 2009. Tracking long-distance songbird migration by using geolocators. Science 323: 896. Tar hi RTON. M K. AND E. Kaiser. 2001. Do fledgling and prebreeding Common Swifts Apus apus take pari in aerial roosting? An answer from a radiotracking experiment. Ibis 143:255-263. Webs t hr, J. D. 1958. Further ornithological notes from Zacatecas, Mexico. Wilson Bulletin 70:243-256. Webster. M, S„ P. P. Marra, S. M. Haig. S. Bensch, and R. f. Hot mes. 2002. Links between worlds: unravel¬ ing migratory connectivity. Trends in Ecology and Evolution 17:76-83. Wiggins, D. 2004. Black Swift ( Cypselnides niger): a technical conservation assessment. USDA. Forest Service. Rocky Mountain Region. Golden. Colorado. USA. www.fs.fed.us/r2/projects/scp/assessments/ blackswift.pdf The Wilson Journal of Ornithology 124(1 ):9- 1 4. 2012 DURATION AND RATE OF SPRING MIGRATION OF KIRTLAND’S WARBLERS DAVID N. EWERT,16 KIMBERLY R. HALL,1 JOSEPH M. WUNDERLE JR.,2 DAVE CURRIE.2 ’ SARAH M. ROCKWELL.4 SCOTT B. JOHNSON/5 AND JENNIFER D. WHITE2’ ABSTRACT. — The duration of migration of the endangered Rutland's Warbler (Setophaga kirtlandii) has not been previously documented. We estimated the average duration of spring migration for five male Kirtland's Warblers by observ ing uniquely color-handed indiv iduals at or near both the beginning and end of spring migration in Eleuthera. The Bahamas, and Michigan, respectively. We estimated the average duration of spring migration for these live individuals to have been no more than 15.8 days (range 13-23 days) and the average distance traveled to have been 144.5 km/day (96. 1 — 169.1 km/day). Received 12 April 2011. Accepted II November 2011. The migratory period is typically the most poorly understood aspect of a migratory species' life history (Faaborg et al. 2010a. b) because of difficulties in studying birds during migration. This information gap constrains our ability lo comprehensively describe population demograph¬ ics, and reduces our ability to effectively imple¬ ment conservation measures (Faaborg el al. 2010a. b) as mortality of adult migratory song¬ birds is apparently high during migration (Sillctt and Holmes 2002). Lack of information on duration of migration, and its relationship to breeding and wintering season conditions, limits our ability lo infer sensitivity of populations to changes in the amount and distribution of habitat required by migrating birds (Faaborg ct al. 20 1 Ob). The challenge of understanding duration of migration, and integrating this information into models of avian population dynamics, is pressing for rare migratory landhirds that are infrequently observed during migration. Models that help explain duration of migration are complex, as many possible intrinsic (c.g.. body condition, experience) and extrinsic (e.g.. weather, habitat distribution, stopover site condi¬ tions) factors influence rates and total duration of The Nature Conservancy. 101 East Grand River Avenue. Lansing. MI 48906. USA. International Institute of Tropical Forestry. USDA Forest Service, Sabuna Field Research Station. MC 02 box 6205, Luquillo, Puerto Rico 00773. Puerto Rican Conservation Foundation, P. O. Box 362495. San Juan. Puerto Rico 00936. Department of Biology. University of Maryland, College Park. MD 20742. USA. Department of Biology. St. Mary's College of Mary¬ land. St. Mary's City. MD 20686, USA. "Corresponding author; e-mail; dewert@tnc.org migration. For example, Cochran and Wikelski (2005) developed a model for Catharus thrushes based on ihe condition of an individual, weather conditions, orientation of flight, and flight dura¬ tion on any given night; they predicted it could take up to 40 days for an individual to travel between the Gulf Coast of Louisiana and its Canadian breeding areas. Estimates of migration duration for passerines have largely been based on extrapolations from banding recoveries of birds en route (Ellegren 1993. Fransson 1995. Newton 2008. Yohannes et al. 2009) rather than documented departure and arrival dates of individual birds from wintering and breeding areas. However, given potential differences in rates of migration along different portions of the route (Stutchbury et al. 2011). duration estimates for banded birds are best obtained over die complete migration (Fransson 1995, Yohannes et al. 2009). New technologies, such as geolocators, now enable researchers to describe the location and duration of migration for individual birds (Stutchbury et al. 2009. 2011; Bachler ct al. 2010, Robinson et al. 2010. Bridge ct al. 201 1. Heckscher et aJ. 2011. Ryder et al. 201 1 ). Use of these techniques is currently limited to species larger than small passerines and, even as the weight of geolocators continues to decrease, their use must be carefully evaluated for imperiled species. Only in rare cases can we estimate duration of an individual’s migration based on observations of birds with unique color-band combinations at the beginning and ending of migration. We report empirical estimates of the duration of spring migration for uniquely color-bunded indi¬ vidual Kirtland's Warblers {Setophaga kirtlandii) observed in the field at the beginning and ending 9 10 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 of migration. The Kirtland’s Warbler, a species designated as federally endangered in the United States and Canada, has a small core breeding range ( — 7 1 km’), and breeds almost exclusively in one distinct habitat type, young stands of jack pine ( Pinas hanksiana) on sandy oulwash soils in the northern Lower Peninsula of Michigan (May- field 1960. Donner et al. 2008 ). Small numbers of birds also breed in similar habitat in Michigan's Upper Peninsula, Wisconsin (Probst et al. 2003, Trick et al. 2008), and southern Ontario (Richard 2008). The total number of singing male Kirt- land's Warblers was estimated to be 1,750 in 2010 (Elaine Carlson, pers. comm.). T his warbler is one of a few passerine species for which it is feasible to search breeding areas for individuals color- banded in wintering areas because locations of all singing males are mapped during an annual population survey (Probst et al. 2005); there arc a relatively small number of birds and sites to check for arriving color-banded individuals. Our objective was to document the duration of spring migration for individual Kirtland’s War¬ blers based on field observations of departure and arrival times of uniquely color-banded birds. To our knowledge, these estimates of spring migra¬ tion duration of Kirtland’s Warblers arc the first derived from observations of the same color- banded individuals in wintering and breeding areas. They also provide comparative data for estimates of migration rates and duration gener¬ ated from other methods. METHODS Teams based in The Bahamas, Michigan. Wisconsin, and Ontario searched the Kirt land's Warbler winter and breeding habitat to estimate duration of migration as part of a coordinated effort to study linkages between winter and summer ranges. We captured 232 Kirtland’s Warblers from 2002 to 2010 in mist nets and color-banded birds at several sites (Wunderle et al. 2010) within 30 km of each other on southern Eleuthera, The Bahamas (—25 N. 76 W; Fig. I Many of the banded individuals were observei repeatedly during a given wintering season. Wi visited Eleuthera sites with several color- bandei Kirtland’s Warblers three to 15 times/season fron mid-April through I May 2003-2010 to documen a date as close to departure as possible. Many o these warMcrs show winter site fidelity (Syke and Clench 1998. Wunderle et al. 2010), bu relocating birds in winter is challenging because some individuals move to different sites, and are difficult to locate and identify in the Lhick shrubby habitats. The Kirtland’s Warbler breeding habitat was searched for color-banded birds from 2003 to 2010. This species often shows breeding site fidelity (Walkinshuw 1983, Mayfield 1992, Bo- cetti 1994; SMR, unpubl. data), like many territorial migrant landbirds. although a few birds disperse to new sites between years (Walkinshuw 1983; DNE and KRH, unpubl. data). We could often estimate arrival dates of individuals return¬ ing to their territories the following spring by checking sites occupied by banded warblers in previous years. Our searches focused on a subset of banded birds that we were able to relocate between mid-to-late April in The Bahamas, for which we had georeferenced their breeding territories in the Lower Peninsula of Michigan in previous years. We monitored territories for a minimum of 30 min/day every I to 3 days from early May through 30 May or until the bird was found. Observers walked through an area encompassing roughly a 200-400 m radius around the georefer¬ enced site during these visits, searching for color- banded birds. Data for birds with known territo¬ ries were supplemented with records of individ¬ uals for w hich we had recorded last observation dates from Eleuthera in mid-to-late April and then opportunistically observed the same bird in Michigan while systematically searching for returning territory holders. Field work was ini¬ tiated each spring, soon after the first confirmed arrival dale of a Kirtland's Warbler in the breeding areas. We calculated each individual's duration of migration, and average distance covered per day from these departure and arrival date estimates. Duration was estimated by calculating the number of nights between the last observation date in The Bahamas (assuming the bird left on the night of the last date it was observed) and the night previous to the first observation in Michigan. The interval between these dates represents the maximum duration of migration. Actual times of migration could be less than the durations reported here because an individual may not have departed immediately following the last Bahamas observation. In addition, individuals may not have been observed on their first day of arrival in Michigan, especially males that did not sing within our sampling period or females. Average Ewert et al. • KIRTLAND'S WARBLER SPRING MIGRATION 11 F-IG. 1. (A) Approximate breeding and wintering locations of color- banded Kirtland’s Warblers for which we estimated duration and distance of spring migration. (B) Breeding locations of color-banded Kirtland's Warblers in the Lower Peninsula of Michigan by county. (C) Approximate location of wintering sites sampled on southern Eleuthera, The Bahamas (described in Wundcrlc et al. 2010). distance traveled/day was calculated by dividing Ihe linear distance traveled between wintering and breeding areas by the number of nights between the last observation of an individual on Eleuthera and the first observation of the same bird in Michigan. Values for both the maximum duration of migration and average distance traveled/day are presented as means ± SD. RESULTS Kirtland’s Warblers were consistently seen on lileuthera through late April 2003-2010 (Wun- derle et al., unpubl. data) with the latest date in any year being 2 May 2006. First arrival dates of males reported anywhere in the Lower Peninsula ‘>1 Michigan ranged from 2 to 7 May during our study and the total number of Bahamas-banded Kirtland’s Warblers observed in breeding areas during the study period in any given year was 7 to 23 (Table 1 ). In comparison, mean arrival dates of males from 2006 to 2010 ranged from 13 to 22 May for the warblers at a subset of Kirtland's Warbler siles in Michigan’s Lower Peninsula (SMR, unpubl. data). The estimated mean duration of migration for five uniquely color-banded male Kirtland’s War¬ blers observed in breeding areas (Fig. 1 ) was 15.8 ± 4.2 days (Table 2). Four of these five birds were observed on the first day the territory was checked and the fifth (band # 2221-08906) was observed 3 days after the territory was last visited. Thus, it is likely the actual mean duration of spring migration for these five individuals was less than the calculated mean of 15.8 days. The mean distance traveled/day, based on this esti¬ mated duration, was 144.5 t 31.2 km. Some observations were not included in our calculations of mean duration of migration because the birds were found opportunistically, and were not detected until relatively late in May 12 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 TABLE I. Number of color-bandcd Kirtland's Warblers from southern Eleuthera. The Bahamas found in breeding areas and dates of first observation of males in the Lower Peninsula of Michigan. Year Number found in breeding areas First observation 2005 8 7 May 2006 12 4 May 2007 23 5 May 2008 14 4 May 2009 16 3 May 2010 7 2 May after the arrival date for most individuals. For example, the range of maximum duration of migration observed for an additional subset of birds, last seen on Eleuthera in mid-to-late April, that were found relatively late in the breeding season in Michigan, was 27-48 days. We also documented that one female, last seen on 27 April 2006 on Eleuthera and first seen in Michigan on 24 May 2006. had a maximum migration duration of 27 days. Estimating arrival dates of females is particularly difficult, even with systematic search¬ es. as they do not sing and may escape detection. DISCUSSION The maximum duration of spring migration for male Kirtland's Warblers from Eleuthera (range = 13-23 days) was shorter than predicted based on models developed for thrushes bv Cochran and Wikelski (2005), but similar to the 13-14 day period for four trans-Gulf spring migrating Wood Thrushes ( Hylocichla mu.stelina ) (Stuchbury el al. 2009). However. Wood Thrushes migrate a longer distance of —3,700 km. compared to -2.200 km by Kirtland’s Warblers. The mean distance traveled/day by male Kin- land's Warblers. 144.5 km. was similar to the spring migration rate of two female Eurasian Hoopoes ( Upupa epops) (122 km/dav and 163 km/day) between the Sahel of western Africa and Switzerland (Baehler et al. 2010) that were tracked with geolocators and within the range of 105-375 km/day traveled during spring migration by a Swainson's Thrush ( Catharus ustulaius » that w-as studied by radiotelemetry between Illinois and Manitoba (Cochran 1987). Daily rates were also within the range of estimates for spring migration of five Sylvia warbler species to Great Britain (range -- 97-232 km/day) and Scandina¬ via (range = 98-163 km/day). based on median capture and recovery dates in the Mediterranean region (Fransson 1995). and for II European passerine species on the European (50-260 kin/ day ) and desert ( 120-150 km/day) portions of the migration route, estimated from median passage rates between Africa and northern Europe (Yo- hannes el al. 2009). One individual fall-migrating Willow Warbler (Phylloscopus trochilus ) traveled 145 km/day (Hilden and Saurola 1982) and another 218 km/day (Hedenstrbm and Peitersson TABLE 2. Estimated duration of migration for color-banded male Kirtland’s Warblers from southern Eleuthera, The Bahamas to known breeding territory locations in the Lower Peninsula of Michigan. Loeations are given in decimal degree coordinates. Individual Last Bahamas observation First Michigan observation and location and location Migration distance (kin) Max duration (days) Migration rate (km/day) 1821-91208 25 Apr 2008 8 May 2008 2.199 13 169.1 24.90787 N 44.4629 N 76.17171 W 84.1809 W 2221-08944 25 Apr 2008 9 May 2008 2.212 14 158.0 24.97589 N 44.4998 N 76.15257 W 83.5794 W 2131-75811 25 Apr 2007 1 1 May 2007 2.222 16 130.7 24.90010 N 44.53544 N 76.14831 W 83.53937 W 2221-08906 24 Apr 2007 17 May 2007 2,211 23 96.1 24.90091 N 44.58206 N 2221-08922 76.15911 W 84.60934 W 27 Apr 2007 24.96263 N 10 May 2007 44.46725 N 2,192 13 168.6 76. 1 7809 W 84.29917 VV Ewert et al • KIRTLAND’S WARBLER SPRING MIGRATION 13 1987). comparable to the migration rates of Kirtland's Warblers sampled. Daily rates of migration by male Kirtland's Warblers were similar to some species, but other species travel at faster rates during spring migration. For instance, four Wood Thrushes migrating from Honduras and Nicaragua to Pennsylvania across the Gulf of Mexico in spring averaged 263 km/day (Stutchbury et al. 2009) and two Purple Martins ( Prague subis ) averaged 281 and 577 km/day during their spring migration from Brazil to Pennsylvania, a distance of —7.550 km. Similarly, five spring migrating Veeries ( Cathants fuscescens) tracked with geo¬ locators. whose estimated migration distance ranged from 5,950 to 10.290 km. migrated faster (209-350 km/day) than Kirtland's Warblers (Heckscher et al. 201 1). and an Aquatic Warbler (Aaocephcilm paliulicola) averaged 280 km/day within Africa during fall migration (Cramp 1992). Overall, however, daily rates of spring migration by these species and our estimates for male Kirtland's Warblers indicate that many near¬ passerines and passerines migrate at rates close to the upper range of migration rates reported in Newton (2008). We obtained estimates of the duration of migration for a small number of males during two spring migration seasons. Ecological, meter ological, physiological, and other factors that affect duration and rate of migration will be better understood (Fransson 1995. kaess 2008. Tottrup et al. 2008, Yohanncs et al, 2009. Stutchbury et al. 201 1) as geolocators and other techniques become available to follow larger numbers of individual birds throughout migration (Robinson et al. 2010. Bridge et al. 2011). It may then be possible to describe other aspects of Kirtland's Warbler migration and connectivity such as distribution of frequently used stopover areas, when and where along the migratory route Kirtland's Warbler are most vulnerable, and inter-seasonal interactions, important hut missing information needed to develop a comprehensive conservation program for the species. ACKNOWLEDGMENTS We thank ihe L'.S. Fish and Wildlife Service. Huron- Manistee National Forest, and Michigan Departmenl of Natural Resources for permission to enter Kirtland's Warbler Management units. The International Program of the L'SDA Forest Service. Michigan Department of Natural Resources, Smithsonian Institution, and The Nature Conservancy funded our work. We greatly appreciate the efforts of Andrew Fra/.ee, P. W. Huber, Samara Lawrentz, Ingeria Miller. Z.oko McKenzie, Ray Perez, Michael Petrucha. Keith Philippe. .1. R. Probsl, Montara Roberts, Robert Slcbodnik, S. .1. Sjogren, Jim Stevens, Mark Thomas, and Jerry Wcinrieh who searched for early- arriving color-handed birds in breeding areas. Joe Fargione, M. L. Herbert. Peter Kareiva. Krista Kirkham, Robert Lalasz. Christopher Rimmer. J. J. Nocera, Peter Weaver, and one anonymous reviewer provided suggestions that greatly improved the manuscript. Sagar Mysorekar pre¬ pared the figure. The work was done in cooperation with the Bahamas National Trust. Puerto Rican Conservation Foundation. University of Puerto Rico, and the Kirtland's Warbler Recovery Team. LITERATURE CITED B \chllr. E.. S. Hahn, M. Schaub, R. Arlettaz. L. Jenni, J. w. Fox. v. Afanasyev, and F. Liechti. 2010. Year-round tracking of smaJl trans-Saharan migrants using light-level geolocators. PLoS ONE 5:e9566. DO 1 : 1 0. 1 37 l/joumal.pone.0009566. BOCETTI. C. 1. 1994. Density, demography, and mating success of Kirtland's Warblers in managed and natural habitats. Dissertation. Ohio State University. Colum¬ bus, USA. Bridge, E. S.. K. Thorip. M. S. Bowlin. P. B. Chilson. R. H. Diehl. R. W. Fleron. P. Hartl. R. Kays, J. F. Kelly, W. D. Robinson, and M. Wikelski. 2011. Technology on the move: recent and forthcoming innovations for tracking migratory birds. BioScience 61:689-698. Cochran, W. W, 1987. Orientation and othet migratory behaviours of a Svvainson's Thrush followed for 1,500 km. Animal Behaviour 35:927-929. Cochran, W, W, and M. Wikelski. 2005. Individual migratory tactics of New World Cathimis thrushes. Current knowledge and future tracking options from space. Pages 274-289 in Birds of two worlds: the ecology and evolution of migration (R. Greenberg and P. P. Marta. Editors), Johns Hopkins University Press, Baltimore. Maryland, USA. Cramp, S. 1992. Handbook of the birds of Europe, the Middle East, and North Africa: birds of the Western Palearctic. Volume 6. Warblers. Oxford University Press. Oxford. United Kingdom. Donnlr, D. M.. J. K. Proust, and C. A, Rime. 2008. Influence of habitat amount, arrangement, and use on population trends of male Kirtland's Warblers. Land¬ scape Ecology 23:467 — 480. Ellec.REN, H. 1993. Speed of migration and migratory Bight lengths of passerine birds ringed during autumn migration in Sweden. Ornis Scandium ica 24:220-228 Faaborg. j., r. t. Holmes, a. D. Anders. K. l. Bildstelv. k. M. Digger, S. a. Gautmeaux Jr.. P. Hegllnd. K. A. Hobso\ a. E. Jaun. D. H Johnson. S. C. Latta. D J. Levey, p. p surra C. L. Merkord, E. Not. S. I. Rows tn\ T. U Sherry. T. S. Sillett. F. R. Thompson III. and N Warnock. 2010a. Consening migrator} land birds in 14 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 the New World: do we know enough? Ecological Applications 20:398^118. Faaborg. J.. R. T- Holmes, A. D. Anders, K. L. Bilustetn, K. M Dugger. S. A. Gauthreaux Jr., P. Heglund. K. A. Hobson. A. E. Jahn, D. H. Johnson, S. C. Latta, D. J. Levey, P. P. Marra. C. L. Mhrkord. E, Noe. S. I. Rothstein. T. W. Sherry. T. S. Sjllett, F. R. Thompson III. and N. WARNOCK. 2010b. Recent advances in under¬ standing migration systems of New World land birds. Ecological Monographs 80:3-48. Fransson. T. 1995. Tinting and speed of migration in north and west European populations of Sylvia warblers. Journal of Avian Biology 26:39-48. Heckscher, C. M.. S M. Taylor, J. W. Fox. and V. Afanasyev. 2011. Veery ( Catharus fuscescens) wintering locations, migratory connectivity, and a revision of its winter range using geolocator technol¬ ogy. Auk 128:531-542. HedenstrOm, A. and J. PbTTERSSON. 1987. Migration routes and wintering areas of Willow Flycatchers Phylloscopus trochilus (L.) ringed in Fcnnoscandia. Ornis Fennica 64:137-143. HildEn, O. and P. Saurola. 1982. Speed of autumn migration ot birds ringed in Finland. Ornis Fennica 59:140-143. Mayfield. H. F. I960. The Kinland’s Warbler. Cranbrook Institute of Science. Bloomfield Hills. Michigan. USA. Mayfield. H. F. 1992. Kirtland’s Warbler (Dendmica kirtlandii). The birds of North America. Number 19. Newton, I. 2008. The migration ecology of birds. Elsevier, London, United Kingdom. Probst, J. R„ D. M. Donnf.r, C. I. Bocetti, and S. Sjogren. 2003. Population increase in Rutland's Warbler and summer range expansion to Wisconsin and Michigan's Upper Peninsula. USA. Oryx 37-365- 373. Probst. J. R.. D. M. Donner. M. Worland, J. Weinrich. P. Hi ber. and K. R. Ennis. 2005. Comparing census methods for the endangered Rutland's Warbler. Journal of Field Ornithology 76:50-60. Raess, M. 2008. Continental efforts: migration speed in spring and autumn in an inner-Asian migrant. Journal of Avian Biology 39:13-18. Richard. T. 2008. Confirmed occurrence and nesting of the Kirtland s Warbler at CFB Petawawa, Ontario: a first for Canada. Ontario Birds 26:2-15. Robinson, W. D.. M. S. Bowlin, I. Bisson, J. Shamoun- Baranes. K. Thorup, R. H. Diehl, T. H. Kinz. S. Mabey. and D. W. Winkler 2010. Integrating concepts and technologies to advance the study ot bird migration. Frontiers in Ecology and the Environ¬ ment 8:354-361. Ryder. T. B.. J. W. Fox, and P. P. Marra. 2011. Estimating migratory connectivity of Gray Catbirds (Dumetclla carolinemi.\) using geolocators and mark recapture data. Auk 128:448-453. SlLI-FTT. T. S. and R. T. Holmes. 2002. Variation in survivorship of a migratory songbird thmughout its annual cycle. Journal of Animal Ecology 71:296-308 Stutchbury. B. J. M.. S. A. Tarof. T. Done, E. Gow. P. M. Kramer. J. Taitin. J. W. Fox. and V. Afanasyev. 2009. Tracking long-distance songbiid migration by using geolocators. Science 323:896. Sitichbury, B. J. M., E. A. Gow. T. Done. M. MAC'PHERSON. .1. W. Fox. AND V AFANASYEV. 2011, Effects of post-breeding moult and energetic condition on timing of songbird migration into the tropics, Proceedings of the Roval Society of London. Series B 278:131-137. Sykes Jr., P. W. and M. H. Clench. 1998. Winter habitat of Kirtland', s Warbler: an endangered Nearctie/neo- tropical migrant. Wilson Bulletin 1 10:244-261. Tottrup, A. P.. K. Thorup, K. Rainio. R. Yosef: E. Ldiiikoine.n, and C. RAHBEK. 2008. Avian migrant' adjust migration in response to environmental condi¬ tions on route. Biology Letters 4:685-688. Trick, J. A.. K. Grveles. D. Ditommaso. and J RoBAIDEK. 2008. The first Wisconsin nesting record ot Rutland's Warbler (Dendmica kirtlandii ). Passen¬ ger Pigeon 70:93-102. Wai.kjnshaw. !.. H. 198.3. Kirtland’s Warbler: the natural htslory of an endangered species. Cranbrook Institute of Science, Bloomfield Hills. Michigan. USA. Wunderle Jr.. J, M.. D. Currie, E. H. Helmer, D. N. F.wert. J. D. White. T. S. Ruzycki. B. Parresol. and C. Kwrr. 2010. Kiriland's Warblers tn anthm- pogenicully disturbed early-successionai habitats ot Eleuthera. The Bahamas. Condor 112:123-137. Yohannes, E.. H. Biebach. G. Nikolaus, and D. J. Pearson. 2009. Migration speeds among eleven species of long-distance migrating passerines across Europe, the desert and eastern Africa. Journal of Avian Biology 40:126-134. The Wilson Journal of Ornithology 124(1): 15-23, 2012 A NEW AREA OF ENDEMISM FOR AMAZONIAN BIRDS IN THE RIO NEGRO BASIN SERGIO H. BORGES1,3'4 AND JOSE M. C. DA SILVA2 ABSTRACT. — We describe a new area of endemism for Amazonian birds which we designate as the Jau Area of Endemism. This area of endemism in central-western Amazonia north of the Rio Solimoes was identified through congruent distributions of six avian taxa: Psophia crepitans ochroptera Pelzeln 1857. Nonnula amaurocephala Chapman 1921. Pteroglossus azara azara Vieilot 1819, Picumnus lafresnayi pusillus Pinto 1936. Synallaxis rutilans confinis Zimmer 1935, and Myrmoborus myotherinus ardesiacus Todd 1927. The southern and eastern limits of this area of endemism are the middle courses of the Solimoes and Negro rivers, respectively. The northern limits apparently coincide with sandy soil vegetation along the middle Rio Negro. The western boundary remains undefined, but could involve the Japura or Iya rivers north of the upper Solimoes. Taxonomic studies and expansion of ornithological collections are needed to more precisely delimit the Jau Area of Endemism. It is possible the avian taxa restricted to the Jau Area of Endemism are derived through parapatric or peripatric speciation events from taxa whose ranges were centered in the Imeri and Napo areas of endemism. Alternatively, tectonic events that affect the lower course of the Rio Negro could influence bird distribution in this region if they serve as vicariance mechanisms. Received 27 June 2007, Accepted 15 July 2011. An area of endemism (AOE) is a "geographical region comprising the distributions of two or more monophyletic taxa that exhibit a phylogenetic and distributional congruence and having their respec¬ tive relatives occurring in other such-defined regions” (Harold and Mooi 1994:261). Areas of endemism are important for at least two reasons. First, they represent the smallest geographical units for postulating hypotheses about the history of their biotas (Cracraft 1988, 1994; Morrone 1994; Morrone and Crisci 1995). Second, these areas are considered priorities for establishment of conservation action because they contain unique biotas (Fjeldsa 1993, Slallerslidd cl al. 1998). There are modern approaches to identifying AOEs including parsimony analyses of endemic- ity (Morrone 1994. Morrone and Crisci 1995, Silva et al. 2004) and optimality criterion (Szumik et al. 2002, Szumik and Goloboff 2004). Tradi¬ tionally. however, these areas have been identified through mapping the congruent geographical distribution of taxa based on extensive area of sympatry but not necessarily complete overlap of distributions (Platnick 1991). Eight AOEs have been recognized for birds in the Amazonian lowlands (Fig. I; Haffer 1974, Departamcnto de Zoolog ia. Museu Paraense Emilio Goeldi. Belem. Para. Brazil. ‘Conservation International do Brazil, Avenue Nazare. 541/sala 310. 66035-170, Belem. Para. Brazil. Current address: Fundayao Vitoria Amazonica, Rua R/ S. casa 07. Quadra Q. Morada do Sol. 69060-080. Manaus. AM. Brazil. 4 Corresponding author; e-mail: sergio@fva.org.br 1978; Cracraft 1985; Silva et al. 2002). Most areas have boundaries coinciding with major rivers of the Amazon Basin (Wallace 1852, Haffer 1978. Cracraft 1985. Ayres and Clutton- Brock 1992). We identified a number of avian taxa during a study of bird species distribution whose ranges apparently are restricted to the lower course of the Rio Negro (Borges 2004a, Borges 2007). Mapping the distributions of these taxa resulted in identification of an AOE not recognized in previous biogeographic anal¬ yses (Haffer 1978, Cracraft 1985). Our objec¬ tive is to describe the new area of endemism based on congruent distributions of six avian taxa. METHODS Study Area. — The study region encompasses the right margin of the lower Rio Negro (Fig. 2). Most specimens analyzed were collected in Jau National Park (JNP), one of the largest (2,272.000 ha) protected areas in Brazil. The avifauna of JNP has been studied for the last 15 years including assessment of species diversity and general eco¬ logical requirements for most bird species (Borges and Carvalhaes 2000; Borges et al. 2001; Borges 2004a. b; Borges and Almeida 201 1). Biogeographic Analysis.— We compiled a list of bird species and subspecies recorded in the study area (Borges et al. 2001, Borges 20()4a, Borges 2007, Borges and Almeida 2011). The taxonomic status and distribution of birds re¬ corded in the study region were evaluated through specimens housed at Museu Paraense Emilio Goeldi (MPEG), Museu de Zoologia da 15 16 THE WILSON JOURNAL OF ORNITHOLOGY • Vo/. 124. No. 1. March 2012 cZZtoJSXSttL SST recosnized for birds in the Amazonia lowla"ds bascd - Haffa ,1974' 1978 Universidade de Sao Paulo (MZUSP), Bird Cc lection of Instituto Nacional de Pesquisas < Amazonia (INPA), Field Museum of Natur History (FMNH), and American Museum Natural History (AMNH). We also compih information on specimens collected in the Negr Japura interfluvium (e.g., Maraa, Barcelos, Man capuru) deposited in MPEG. All relevant inform tion available from the taxonomic literature w; also used (Pinto 1944. 1978; Sherman 199- Winkler and Christie 2002; Borges 2004a). W did not attempt a complete and detailed taxonom revision of the bird taxa. but checked if the specit °heir cTof'f T" m°tphologically distinct fro, ,heir ranges ™ The new area of endemism is at proximities r endemic to the Napo and Imeri areas of endemism as originally proposed by Haffer (1978) and Cracraft (1985). RESULTS The taxonomic and geographical distributions ol 383 bird species and subspecies were assessed in the study region. The geographic distribution of one species, Chestnut-headed Nunlet (Nonnula amaurocephala), and five subspecies, Grey¬ winged Trumpeter (Psophia crepitans ochropierdl Ivory-billed Aracan ( Pteroglossus azora azQto)- Lafresnayes’s Piculet (Picumnns lafresnayi pail- lus). Ruddy Spinetai! (SynalUixis rutilims confinis). and Black-faced Antbird ( Myrrnoboms mother- inns ardesiacus) are restricted to the west of the lower Rio Negro anil north of the Rio Solimocs. I hese taxa are considered endemic to the study region. Borges and Silva • A NEW AREA OF ENDEMISM FOR AMAZONIAN BIRDS 17 «8WW TW*V Tabocal Barcelos Muirapinina Japura ^ Manaus Igarape >— Cacau / Pereira Manacapuru ( CodajSs HMl3 Castanheiro Santa Maria Acajatuba FIG. 2. Northwestern Amazonia, showing the study region and municipalities cited in the text. The light gray area indicates the suggested limits of the Jati Areu of Endemism. Grey-winged Trumpeter (Psophia crepitans ochroptera Pelzeln 1857). Pcl/.eln described this trumpeter as a distinct species. Psophia ochrop¬ tera. Ochroptera is treated as a subspecies of P. leucoptera in recent taxonomic literature (Pinto 1978. Sherman 1996). We agree with Haffer (1974) who proposed that ochroptera is a subspecies of P. crepitans , the trumpeter species distributed north of the Amazon River. P. crepitans has two other subspecies in addition to ochroptera: crepitans inhabiting the Guianan AOE and napensis. inhabiting the Napo and Imeri AOEs. P. c. ochroptera has been recorded mostly in white-sand woodland and terra finite forest in JNP (Borges et al. 2001, Borges and Almeida 701 1 ). where an adult female was collected (INPA # 576). A recent molecular systematic study i Ribas et al. 201 1 ). in addition to morphological assessment (Haffer 1974). found that ochroptera is diagnosable from the other subspecies in the Psophia crepitans complex. Chestnut-headed Nunlet (Nonnula anmuroce- phula Chapman 1921). This rare puffbird was known only from specimens collected in Mana¬ capuru used in the original description until it was rediscovered in JNP (Whittaker et al. 1995). It is associated with seasonally flooded black- water forest or Igapo forest (Whittaker et al. 1995, Borges et al. 2001). The MZUSP has skins collected in Manacapuru (MZUSP 16561) and Codajas (MZUSP 16387), and one specimen was collected in JNP (MPEG 55855) in addition to specimens mentioned in Whittaker et al. (1995). The northernmost record of N. amaurocephala is the Unini River (Whittaker et al. 1995) and the easternmost is the Amana River (M. Cohn-Haft, pers. comm.) with no confirmed records from the upper portion of the Rio Negro (Haffer and Fitzpatrick 1985). It is suggested that N. amaurocephala forms a superspecies with N. ruficapilla and N. frontalis (Rassmussen and Collar 2002) Ivory-billed Aracari {Pterogtossus azara azara Vieilot 1819). This subspecies of the Ivory-billed Aracari has the upper mandible brownish colored and is recorded from Castanheiro. Codajas. Igarape Cacau Pereira (Haffer 1974). Manacapuru (MZUSP 16833 and 16834), Maraa (MPEG 18 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. I. March 2012 42505-42507), and JNP (Borges et al. 2001, Borges 2007). It is replaced by P. a. flavir osiris in the Napo and Imeri AOEs and by P. a. marine south of the Solimdes (Haffer 1974). There is some evidence of hybridization between marine and azara near the mouth of the Purus River (Haffer 1974: 222-223). This Aracari has been recorded in both terra finne and Hooded forests in JNP (Borges et al. 2001. Borges and Almeida 201 1). Lafresnaye's Piculet (Picumnus lafresnayi pusillus Pinto 1936). P. pusillus was described from Codajas on the left margin of the Rio Solimdes. by Pinto (1936). This taxon was later reassigned as a subspecies of P. aurifrons (Peters 1948). Recent taxonomy considered pusillus as a subspecies of Picumnus lafresnayi (Winkler et al. 1995, Winkler and Christie 2002). P. lafresnayi forms a superspecies with P. puniilus that is apparently restricted to southern Venezuela and northwestern Brazil (Winkler and Christie 2002). There are six MZUSP specimens identified as P. I. pusilus collected in the Manacapuru region (MZUSP 16614-16619). The plumage of a single speci¬ men collected in JNP and deposited in the INPA Bird Collection matched better with P. /. pusillus than with P. puniilus. The geographical distribution of pusillus appears to be restricted to the region west of the lower Rio Negro being replaced by P. puniilus in the Imeri AOK (Fig. 2). P. I. pusillus has been found mainly in Hooded forests and secondary growth in 20 fl )(B°rgeS Ct al' 2°01, Borges and Almeida Ruddy Spinetail ( SynaJlaxis rutihms confin Zimmer 1935). This subspecies was describe from specimens collected in Igarape Caca Pereira in the lower portion of Rio Negro. It , replaced by S. r. dissors on the left margin of Ri Negro, which has been found in Manaus and i the upper Rio Negro at Sao Gabriel da Cachoeir and along the Casiquiare channel (Friedman 1948 Pinto 1978). A better delimitation of th distribution of S. r . eonfinis will require specime collections south of die Negro and Uaupes riven p '• “>,ir,,,is is replaced in Napo AGE (easier Ecuador and northeastern Peru) by ,V. r. caque south onhe's AOE ,upper by * r- (ms) LFnTZ moborus myotherinus and concluded that arde- siacus was a diagnosable taxon This subspecies was described from Manacapuru and has been recorded in Igarape Cacau Pereira. Santa Maria. Codajas (Zimmer 1932, Pinto 1978), Mania (MPEG 42654-42672), and Jau National Park (MPEG 50614—50620). and is replaced in Imeri and Napo by M. tn. elegans and in Inambari by M. m. myotherinus (Haffer and Fitzpatrick 1985). This antbird has been recorded in mature upland forest, white-sand campinaranas, and secondary growth forest in JNP (Borges et al. 2001. Borges and Almeida 20 1 I ). Another seven avian taxa also are apparently restricted to centra I -western Amazonia: Anuizona autumnaUs diadema. Braehygalbula lugubm phaeonota, Nunnula rubecula simidatrix, Hyki estastes stresemanni stresemanni . Sc! e runts rufi- gularis hrunnescens, Myrmoborus lugubris Stic- top terns, and Hylophylax naevius ohscunts. These subspecies, however, are known only from a few specimens and their taxonomic status and distri¬ bution needs additional evaluation. Napo and Imeri Areas of Endemism.— The Imeri and Napo areas of endemism were supported by 21 and 56 avian taxa as originally described (Haffer 1978. Cracraft 1985). An updated distributional and taxonomic assessment decreased these num¬ bers to six species (or subspecies) endemic to linen and 42 endemic to the Napo AOE (Appendix). Several bird species previously thought to be restricted to the upper reaches of the Rio Negro have had their geographical distribution extended southward and to northern Peru (Borges et al 2001, Alonso and Whitney 2003, Borges and Almeida 201 I ). Other Imeri birds are restricted to the northwestern portion of the Amazon Basin with no confirmed records in the lower reach ol the Rio Negro or northern Peru. Most species restricted to Imeri AOE are specialists in white sand campinas (Cyanocorax helprirti and Myrvu- eiza pelzelni ), montane and pre-montanc forests ( Percnostola caurensis), and flooded forest (77m pophaga cherriei) (Zimmer 1999; Hilty 2003: SMB. unpub], data). Similarly, some birds previously thought to be endemic to the Napo region had their distributions extended east to the Rio Negro, as in the case of My mint hernia ignota and Pteroglossus pluricinc- tus (Borges and Almeida 201 1). Some species are even more widespread than previously described, but there are species endemic to northwestern (Imeri endemics), upper north Amazon (Napo Borges and Silva • A NEW AREA OF ENDEMISM FOR AMAZONIAN BIRDS 19 endemics), and the central-western portion of the basin (this study). DISCUSSION We describe a previously undesignated area of endemism in central-western Amazonia that is supported by the congruent distribution of six avian taxa. Haffer and Fitzpatrick (19X5) noted that a small number of strongly differentiated bird populations were restricted to this region but did not recognize it formally as an AOE due the reduced number of avian taxa (J. Haffer, pers. comm.). Conceptually and operationally, however, an AOE could be recognized with just two species (Platinick 1991, Harold and Mooi 1994). Amorin and Pires (1999) also identified a small AOE bounded by the Rio Solimoes and Rio Negro without offering information about which species supported this biogeographic component (Amorim and Pires 1999: fig. 27). We suggest naming this new area of endemism as Jau in recognition of the importance of the National Park to biodiversity conservation in the Amazon. This area appears nominated as Rio Negro Area ol Endemism in previous publications (Borges 2007, Ribas et al. 2011). However, we believe that Jau Area of Endemism is more appropriate to the geographical setting of the area (Fig. 2). The precise geographical limits of this area are only partially identifiable. The southern and eastern limits probably coincide with middle portions of ihe Solimoes and Negro, large rivers known to isolate bird populations (Sick 1967, Haffer 1992). The northern boundary is more complicated, but from the middle Rio Negro northward the landscapes dominated by terra Jirme forests are extensively replaced by forests and fields growing in sandy soils called 'campi- naranas' and 'campinas' (Anderson 1981. IBGE 1997), This more open vegetation occupies thousands of square kilometers between the middle and the upper Rio Negro. It is possible this discontinuity in vegetation serves as a barrier (or filler) to dispersal for some bird species, especially those that occur more frequently in terra firme forest on clay soils. The western boundary of Jau AOE also is difficult to identify due to the scarcity of bird collections from the region between Maraa to the Brazilian-Colombian border. We tentatively suggest the courses of the Japura or l^a rivers as the western boundary of Jau AOE, although additional fieldwork will be necessary (Fig. 2). We also note this region of faunal turnover could be not coincident with any river course as happens in the southern portion of the Amazon Basin (Haffer 1992). One fundamental concern in recognizing and delimiting the Jau AOE is the current taxonomy of the birds considered endemics. The use of subspecies in biogeographic analyses is problematic because an unknown number of bird taxa described in the ornithological literature are not discrete evolutionary units (Cracraft 19X5, Haffer 1987). Some authors have been successful in using subspecies for biogeographic analysis (Bates et al. 1998, Borges 2007). Analyzing polytypical species as a single entity tends to over-estimate the geographical distributions of taxa and potentially ignores relevant units for conservation and bioge¬ ography (Bates and Demos 2001). We considered only taxa with accentuated morphological differen¬ tiation. However, the taxonomic status of Amazo¬ nian bird species, including those discussed here, needs to be continuously evaluated and their phylogenetic relationships clearly established. These taxonomic and biogeographic studies will require additional ornithological collections in the northern portion of the upper Rio Solimoes and in the middle to upper Rio Negro, principally along the right margin ol' the latter, as these regions are likely contact zones. The phylogenetic relationships of the species and subspecies endemic to the Jau AOE are not adequately known. However, three taxa (Pter- oglossas azara azara. Picummis lafresnayi pusil- Iti.s, and Mynnoboriis myotherinus artlesiacus) appear to have their closest relatives in Ihe lmcri and Napo AOEs. This is supported by Borges (2007), who applied quantitative methods (parsi¬ mony analysis of endemicity and cluster analyses) to analyze the biogeographic relationships be¬ tween avifaunas from different Amazonian AOEs. The avifauna of the JNP shares more species and subspecies with localities in the Imeri and Napo AOEs than with localities in the Inambari and Guiana AOEs (Borges 2007). This suggests the barrier effect of the Negro and Solimoes rivers is stronger than the barrier effect in the western border ol the Jau AOE. A working hypothesis to the evolutionary history' of the six taxa restricted to the Jaii AOE is that they wea* derived through parapatric or peri panic speciation from ancient species whose ranges were once centered in the Imeri and \apo AOEs. Alternatively , tectonic events that affect the lower course of the Rio 20 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 Negro (Forsberg et al. 2000. Almeida-Filho and Miranda 2007, Silva et al. 2007) could influence bird distribution in this region if they serve as vicariance mechanisms (Ribas et al. 2011). We call attention to the urgent need to review the currently recognized AOEs for Amazonian birds. The last review of these areas w as compiled over 20 years ago (Cracraft 1985), and taxonomic and geographical distribution data have continued to accumulate. The areas of endemism for Amazonian birds should be re¬ evaluated through modem biogeographic approaches (Morronc and Crisci 1995, Szumik and Goloboff 2004) facilitated by continuous progress increasing the data base of geographical distribution and systematics of neotropical birds. ACKNOWLEDGMENTS SHB acknowledges Capes. WWF-Brazil, and Fundajao Vitdria Ama/onica for financial and technical support. We thank Brazil's federal agency of environmental protection (IBAMA) for permission to work in Jaii National Park. Luis Fdbio Silvcira, Maria Luiza. Alexandre Aleixo, and Mario Cohn-Haft kindly permitted us to examine bird collections under their care, Marcelo Moreirn kindly prepared the maps. Morton Isler, Charles Zartman. and Camila Ribas helped in first versions of the paper. We appreciate the improvements in English usage by Pltil Stouffcr through the Association of Field Ornithologists' program of editorial assistance. We are also grateful to J. M. Bates and Jurgen Haffer for careful reviews of this paper and useful suggestions. We dedicate this paper to the memory of Dr. Jurgen Haffer for his inestimable contribution to the biogeography of neotropical birds. LITERATURE CITED Almeida-Filho. R. and F. P. Miranda. 2007. Mega capture of the Rio Negro and formation of the Anavilhanas Archipelago, central Amazonia, Brazil: evidences in an SRTM digital elevation model. Remote Sensing of Environment I 10: 387-392. Alonso, J. A. and B. M. Whitney. . * . distributional records of birds from white-sand fores of the northern Peruvian Amazon, with implicatio for biogeography of northern South America. Cond 105:552-566. Amorim. D. S. and M. R. Fires. 1999. Neotropic biogeography and a method for maximum biodiversi estimation. Pages 183-219 in Biodiversity in Brazil: first approach- Proceedings of die workshop met hoc tor the assessment of biodiversity in plants ar E M. Bicudo and N. A. Meneze Editors). CNPq, S3o Paulo, Brazil AN°aT' A' T' Whi,C '“nd "V-*" of Brasilia Amazonia. Biotropica 13:199-210 Ayres, M and T. h. Clutton- Brock. 1992 RiVt nrim'u'nCSA NpC^('CS ranSe in Amazonia primates. American Naturalist 140: 531-537 Bates, J. M. and T. Demos. 2001. Do we need to devalue Amazonia and other large tropical forests9 Diversity and Distributions 7:249-255 Bates, J. M., S. HacketT. and J. CRACRAIT. 1998. Area- relationships in the neotropical lowlands: a hypothesis based on raw distributions of passerine birds. Journal of Biogcography 25:783 793. BORGES. S. H. 2004a. Avifauna do Parque National do Jaii: urn estudo integrado cm biogeografia, ccologia de paisagens c conservayao. Dissertation. Muscu Para- ense Emilio Goeldi, Paid, Brazil. BORGES. S. II. 2004b. Species poor but distinct: bird assemblage in white sand vegetation in Jaii National Park, Brazilian Amazon. Ibis 146:114-124. BORGES, S. H. 2007 Analise biogeografica da avifauna da rcgiiio oeste do baixo Rio Negro. Amazonia Brasilcira. Revista Brasilcira dc Zoologia 24: 919-940. Borges, S. H. and R. A. Almeida. 201 1. Birds of the Jau National Park and adjacenl area, Brazilian Amazon, new species records with reanalysis of a previous checklist Revista Brasilcira de Omitologja 19:108-133. Borges, S. H. and A. M. P. Carvaluaes. 20(X). Bird species of black water inundation forests in the Jau National Park (Amazonas State. Brazil): their contn- bulion to regional species richness. Biodiversity and Conservation 9:201 214. Borges, S. H.. M. Cohn-Haft, A. M. P. Carvaluaes, I.. M. Henriques, j. f. Pacheco, and A. Whittaker, 2001. Birds of Jaii National Park, Brazilian Amazon: species check-list, biogeography and conservation. Omitnlogiu Neotropical 12: 109-140. Cohn-Haft, M„ A. Whiitaker, and P. C. Sioltfer. 1997. A new look at the "species-poor" central Amazon: the avifauna north of Manaus, Brazil Ornithological Monographs 48:205-235. Cracraft. J, 1985, Historical hiogeography and patterns of differentiation within the South American avifauna: areas of endemism. Ornithological Monographs 36:49-84. Cracraft, J. 1988. Deep-history biogeography: retrieving the historical pattern of evolving continental biotas. Systematic Zoology 37:221-236. Cracraft, J. 1994, Species diversity, biogcography, and the evolution of biotas. American Zoologist 34:33-4”. Fitzpatrick, J, W. 2004. Family Tyrannidae. Pages 1 70— 463 in Handbook of the birds of the world. Volume 9. Cotingas to pipits and wagtails ( J. del Hoyo, A. Elliott, and D. A. Christie. Editors). Lynx Edicions. Baac- lona. Spain, FjeldsA, J 1993. The avifauna of the Polxlepsis woodlands of the Andean highlands: the efficiency of basing conservation priorities on patterns of endemism. Bird Conservation International 3:37-55. Forsberg. B. R . Y. Hashimoto. A. Rosenqvist. and F. P Miranda. 2000. Tectonic fault control of wetland distributions in the central Amazon revealed by JERS-1 radar imagery. Quaternary International 72: 61-66. FRIEDMANN, H. 1948. Birds collected by the National Geographic Society 's expeditions to northern Brazil and southern Venezuela. Proceedings of the United States National Museum 97:373-570. Borges and Silva • A NEW AREA OF ENDEMISM FOR AMAZONIAN BIRDS 21 Gill, F. and D. Donsker (Editors). 201 1. IOC World bird names. Version 2.9. Princeton University Press, Prince¬ ton. New Jersey, USA. http://www.worldhirdnynieN.org/ Glilherme, E. and S. H. Borges. 2011 Ornithological records from a campina/canipinarana enclave on the upper Jurua River, Acre. Brazil. Wilson Journal of Ornithology 1 23:24 — 32. Haffer, J. 1974. Avian speciation in tropical South America. Publications of the Nuttall Ornithological Club 14:1-390. Haffkr. J. 1978. Distribution of Amazon forest birds. Bonner Zoologischc Beitraege 29:38-78. Haffer. J. 1987. Biogeography of neotropical birds. Pages 105-150 in Biogeography and quaternary history in tropical America (T. C Whitmore and G. T- Prance, Editors). Clarendon Press, Oxford, United Kingdom. Haffer, J. 1992. On the "river effect'' in some forest birds of southern Amazonia. Bolctim do Museu Paruense Emilio Goeldi (Seric Zoologia) 8:217-245. Hafffr. J. 1997. Contact zones between birds of southern Amazonia. Ornithological Monographs 48:281-305. Haffer. J. and J. W. Fitzpatrick. 1985. Geographic variation in some Amazonian forest birds. Ornitho¬ logical Monographs 36:147 168. Harold, A. S. and R D. Moot, 1994. Areas of endemism: definition and recognition criteria. Systematic Biology 43:261-266. HtLTY, S. L. 2003. Birds of Venezuela. Second Edition. Princeton University Press, Princeton. New Jersey, USA. Hu. D. S„ L. Joseph, and D. Agro. 2000. Distribution, vanatioti, and taxonomy of Topaui hummingbirds ( Aves: Trochilidae). Ornitologia Neotropical 1 1:123-142. 1BGE. 1997. Diagnostics Ambiental da Amazonia Legal (CD-ROM). IBGE, Rio de Janeiro. Brazil. Lane, D. F.. T. Vai.qui, j. A, Alonso.. J Armenta, and K. Eckiiardt. 2006. The rediscovery and natural history of the White-masked Antbird (Pilhys casta- neus). Wilson Journal of Ornithology I 18:13-22. Morrone, J. J. 1994. On the identification of areas of endemism. Systematic Biology 43:438 441. Morrone, J. J. and J. V. Crisci. 1995. Historical biogeography: introduction lo methods. Annual Re¬ view of Ecology and Systcmalics 26:373-401. Naka. L. N.. M. Cohn-Haft. F. Mai iht- Rodrigues, M. P. d. Santos, and M. f. Torres. 2006. The avifauna of the Brazilian State of Roraima: bird distribution and biogeography in the Rio Branco Basin. Revista Brasileira de Ornitologia 14: 197-238. Payne, R B. 1997. Family Cuculidae (Cuckoos). Pages 508-607 in Handbook of the birds of the world. Volume 4. Sandgrouse to cuckoos (J. del Hoyo, A. Elliott, and D. A Christie. Editors). Lynx Edicions. Barcelona. Spain. Peters. J. L. 1948. Check-list of birds of the world. Volume 6. Museum of Comparative Zoology. Harvard University. Cambridge, Massachusetts. USA Peters. J. L. 1951. Check-list of birds of the world. Volume 7. Museum of Comparative Zoology , Harvard University, Cambridge. Massachusetts, USA. Pinto. O. 1936. Notas de ornithologia amazonica. Revista do Museu Paulista Tomo 20:229-244. PINTO, O. 1944. C’aialogo de aves do Brasil (Second Part). SecretSria da Agriculture, Industria e Comercio, Sao Paulo. Brazil. Pinto, O. 1978. Novo catalogo das aves do Brasil (First Pari). Suo Paulo. Brazil. Pt.ATiNICK. N. I. 1991. On areas of endemism. Australian Systematic Botany 4:xi-xii. Rasmussen. P. C. and N. J. Collar. 2002. Family Bucconidae (Pullbtrds). Pages 102-138 in Handbook of the birds of die world. Volume 7. Jacamars to woodpeckers (J. del Hoyo. A. Elliott, and D. A. Christie, Editors). Lynx Edicions. Barcelona, Spain. Restall. R.. C Ropner. and M. Lentino. 2006. Birds of northern South America-an identification guide (2 volumes). Yale University Press. New Haven. Con¬ necticut. USA and London. United Kingdom. Ribas. C. C., A. Auuxo. A. C. K. Nogueira, C. Y. Myiaki, and J. Cracraet. 20 1 1 . A pulaeobiogeographic model for biotic diversification within Amazonia over the past three million years. Proceedings of the Royal Society of London, Senes B. DOl: 10. 1 098/rspb.20l 1. 1 1 20. RlDGELY. R. S. AND G. Ti dor. 1989. The birds of South America. Volume I. The oscine passerines. University of Texas Press, Austin. USA. Robbins. M. B.. M. J. Braun. C. J. Huddleston, D. W. Finch, and C. M. Mieensky. 2005. First Guyana records, natural history und systematics of the White- naped Seedcater Dolospingus fringilloides. Ibis 147:334-341. SCHUCHMANN, K. L 1999. Family Trochilidae (Humming¬ birds). Pages 32 108 in Handbook of the birds of the world. Volume 5. Barn-owls to hummingbirds (J. del Hoyo, A. Elliott, and D. A. Christie, Editors). Lynx Edicions, Barcelona. Spain. Sherman. P. T. 1996. Family Psophiidae (Trumpeters). Pages 96-107 in Handbook of the birds of the world. Volume 3. Hoatzins to auks (J. del Hoyo, A. Elliott, and J. Sargatal. Editors). Lynx Edicions, Barcelona, Spain. SlC’K, H. 1967. Rios e enchentes na Amazonia como obst&culo para a avifauna. Atas Simposio sobre a Biota Amazonica 5: 495-520. Silva. C. L., N. Morales. A. P. Crosta, S. S. Costa, and J. R. Jimfnez-Rueda. 2007. Analysis of tectonic- controiled fluvial morphology and sedimentary pro¬ cesses of the western Amazon Basin: an approach using satellite images and digital elevation model. Anais da Academia Brasileira de CiSncias 79- 693-711. Silva. J. M. C„ F. C. Novaes, and D. C. Oren. 2002. Differentiation of Xiphocolaptes (Dendrocolaptidae) across the river Xingu. Brazilian Amazonia: recogni¬ tion of a new phy logenetic species and biogeograph- ical implications. Bulletin of the British Ornithological Club 1 22:1 85- J 94. Silva. J. M. C., M. C. Souza, and C. H. M Castellfth. 2004. Areas of endemism for passerine birds in the Atlantic Forest. South America. Global Ecology and Biogeography 13:85-92. 22 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 Snow, D. W. 2004. Family Cotingidae (Cotingas). Pages 32-108 in Handbook of the birds of the world. Volume 9. Cotingas to pipits and wagtails (J. del Hoyo. A. Elliott, and D. A. Christie, Editors). Lynx Edicions, Barcelona, Spain. Statterseield, A. J.. M. J. Crosby. A. J. Long, and D. C. Wege. 1998. Endemic bird areas of the world: priorities for bird conservation. BirdLife International, Cambridge. United Kingdom. Szumik. C. and P. Gogoboff. 2004. Areas of endemism: an improved optimality criterion. Systematic Biology 53:968-977. Szumik, c„ f. Cuezzo. P. Goloboff, and A. Ciiai.up. 2002. An optimality criterion to determine areas of endemism. Systematic Biology 51:806-816. Tobias. J A.. T. Zuchner, and T. A. Melo-Junior. 2002. Family Galbulidae. Pages 296-555 in Handbook of the birds of the world. Volume 7. Jacamars to woodpeck¬ ers (.1. del Hoyo. A. Elliott, and D. A. Christie. Editors), l ynx Edicions, Barcelona, Spain. Wallace. A. R. 1852. On the monkeys of the Amazon. Proceedings ol the Zoological Society of London 1852: 107-110. Whittaker. A.. A. M. P. Carvalhaes. and J. F. Pacheco. 1995. Rediscovery of the Chestnut-headed Nunlet Nonnula atmurocephala in Amazonian Brazil Cotinga 3:48-50. W inkier. H. AND D. A. Christie. 2002. Family Picidac (Woodpeckers). Pages 296-555 in Handbook of ihc birds of the world. Volume 7. Jacamars to woodpeck¬ ers (J. del Hoyo, A. Elliott, and D. A. Christie, Editors). Lynx Edicions. Barcelona. Spain. Winkler. H., D. a. Christie, and D, Nurney. 1995 Woodpeckers a guide to the woodpeckers of the world. Houghton Mifflin Company. Boston. Massa¬ chusetts, USA. Zimmer. J. T. 1932. Studies of Peruvian birds. IV. The lormicurian genera Myrmoborus and Mvmecizti in Peru. American Museum Novitates 545:1-24. Zimmer, K, J. 1999. Behavior and vocalizations of the Caura and the Yapacana anthirds. Wilson Bulletin 1 1 1:195-209. Zimmer, K. J. and M. I Islkr. 2003. Family Thamnophi- lidae (Typical antbirds). Pages 448-681 in Handbook of the birds of the world. Volume 8 (J. del Hoyo, A. Elliott, and D. A. Christie, Editors). Lynx Edicions. Barcelona. Spain. APPEMDEX. Distribution of avian taxa in Imeri and Napo areas of endemism as originally proposed bv Haffer (1978. ss “h — »ith c>- ^ -- «">■ Original name Crypturellus casiquiare — H C. duidae — C Aramides calopierus — H Mini sal vini — H & C M. tomentosa — H Neomorphus pucheranii — C C ampv lop terns villa viscensio — C Leucippus chlorocercus — H & C Heliodoxa gularis — H & C Topaza pyra — H & C Phlogophilus hemileucurus Pyrrhura albipectus — H & C Galbula tombacea — H G. leucogastra chalcothorax G. pastazae — H Gal ba I cyrhynch us leucotis — H Nonnula ruficapilla rufipectus — C N. brunnea — H Notharchus ordii — H Pteroglossus f. flavirostris — H P • pluricinclus — H Selenidera r. reinwardtii — H & C S. nattereri — H & C Picumnus pumillus- — H & C Celeus spectabilis — H & C Hylexetastes stresemanni insignis— Synallaxis moesta — H & C Thripophaga cherriei — H & C Neoctanies niger ■ — H Current name Imeri Napo Jad1 Crypturellus casiquiare 2 C. duidae •' A ramides calopierus Mini salvini M. tomentosum Neomorphus p. pucheranii 4 C ampyiopterus villaviscensio Leucippus chlorocercus Heliodoxa gularis Topaza p. pyra 5 Phlogophilus hemileucurus Pyrrhura albipectus Galbula tombacea tombacea 6 G. chalcothorax’ G. pastazae Galbal cyrhynchus leucotis Nonnula ruficapilla rufipectus N. brunnea Notharchus ordii * Pteroglossus azara flavirostris P pluricinclus Selenidera r. reimvardtii S. nattereri Picumnus pumillus Celeus spectabilis spectabilis Hylexetastes stresemanni insignis Synallaxis moesta Thripophaga cherriei Neoctanies niger'’ x x E E x E E E x E E E E E E E E x X X E Borges and Silva • A NEW AREA OF ENDEMISM FOR AMAZONIAN BIRDS 23 APPENDIX. Continued. Original name Current name Imeri Napo Jau* Myrmotherula sunensis — H & C Myrmotherula sunensis sunensis10 E M. obscura — H M. ignota obscura" X X X A/, longicauda soderstromi — C M. longicauda soderstromi E V/. cherriei — H & C M. cherriei'1 X X X V/ ambigua — H & C M. ambigua X X Herpsilochrius stictunis dugandi — H & C Herpsilochmus dugandi E H. dorsimaculaius — H H. dorsimaculaius 13 X X Hypocnemis hypoxunthu — C Hypocnemis h. hypoxantha 14 X X X Pitlivs casianea — C Pithys castaneus ,s E Rltegmuturhinu m. melanosticta — C Rhegmatorhina m. melanosticta E R. iris lata — C R. cristata X X Mynneciza rnelanoceps — H Mynneciza rnelanoceps16 E M. pelzelni— H & C M. pelzelni E M. ilisjuncta — II & C M. disjuncta'1 X X Gymnopilhys leucaspis casianea — C Gymnopilhys leucaspis castaneus E G. leucaspis late mil is — C G. leucaspis lateralis X X Thamnophius praecox — H & C Thamnophius praecox E Pe re nos tola cau rensi v — H Schistocichla caurensis 18 E Grallaria Jigidssima — H & C Grallaria dignissima E G. futviveniris — H & C Hylopezus f fulviventris E Pipra coronata coronata — H Lepidotrhix coronata coronata E P. coronata caquetae — C L coronata caquetae E P. pipra discolor — C Dixiphia pipra discolor E Heterocercus aurantiivertex — H & C Heterocercus aurantiivertex E H. Jlavi vertex — II & C H. flavivertex 19 X X Attiia ciiriniventris — H & C A ttila citrini ventris20 X X X A. torridus — C A. torridus E Myiophobus cryptoxantluis — H & C Myiophobus cryptoxantluis E Ramphotrigon fuscicauda — H Ramphotrigon fuscicauda1' E Todirostnnn c. ctdopterum — H & C Poeciiotriccus culoptertts E /. c. capiiale — H P. capitalis 11 E Contopus n. nigrescens — C Contopus n. nigrescens E M icrnbales cinereivennis — H Microbates cinereiventris hormotus 23 E Cyphorimis arudus sidvini — C Cyphorinus aruda sulvini E Microcerculus bamhla albigularis — C Microcerculus bamhla albigularis14 E Cacicus sclateri — II & C Cacicus sclateri E Ocyatus latiroslrix—C Ocyalus latirostris E Hylophilus hypoxanlhus fuscicapiUus- C llylophilus hypoxanthus fuscicapiUus E //. b. brunneiceps — C H. brunneiceps 16 X X Cyanocarux heilprini — H & C Cyanacora. \ he Up tin i E Dolnspingus fringilloides — H & C Dnlospingus fringilloides 26 X X Ia"wmic and distributional noles 1 1 1 Kefen-mcs lot species records in the )aii Area of Endemism were Borges e! al. (2001) and Borges and Almeida (201 1). I- 1 Recently recorded in northern Peru (Alonso and Whitney 2001), (}) Recently recorded in northern Pern (Alonso and Whitney 2003.1. (4) Subspecies N. p. :-;'i,l„phanei occurs south of Kio Soli miles (Payne 1997). Ilaffcr I 1997) reported ,V. p. pucherani tn rhe upj>er Rio Negro (Intern. (5) Recorded in sou- ihcni Venezuela i Hilly 2003 1. (6i Subspecies G i mrnitilh recorded south of Rm Solimfle* (Inambari) (Tobias el al. 2002). (7) Also recorded in (he Jurud River iln.unhinl i I'obias cl al. 2002). ( Kl Recently recorded in Inamhan AOE (Guilhenne and Borges 201 1 r. id) Records from Inambari. Imcri. and Tapajos AOEs 1 Zimmer and Isler 2003) ( lOl Subspecies M >. mvu pi recorded in Junta River < Inambari l (Zimmer and Isler 2003). (II) Also recorded in Inambari AOE (Zimmer • n.l Isler 2001) r 12) Recently recorded in rtWtliem Hem (Alonso and Whilney 2003). 1 1 3 1 Also recorded in Guiana AOE (Cohn-Hafi et a). 1097. Zimmer and Isler 2**03. Naka et al 2006). 1 14) Also irvm.lcil in Inambari AOE (Zimmer and Isler 200.3). 1 15) Recently rediscovered in field (Lane el al. 2006). I ten Also recorded in Inambari AOE (Zimmer and Isler 2003 1. 1 7 1 Recently recorded in Guiana AOE (Naka cl al. 2006). 1 18) Recently rediscovered in field (Zimmer 1999). (|9) Also recorded in Guiana AOE (Snow 2004 Naka el al 2006). i20) Recently recorded in Inamhan AOE (Guilherme and Borges 2011). (21 1 Considered monofypic *>lf) several records south of (lie Amazon, is River w ith an isolated population in the .Vapo region (Fil/pairick 2004). (22) Considered mondlypic with a population recorded in Kondoniu AOE (FiUpafrick 20(4) (23) Dislribulton and taxonomy follow Resell el al. (2006). (24/ Distribution and latonomv follow Rcsiall (2006) (25 1 Considered as tnonoiypte by Ridgely and Tudor (198V) (2 6) Recently recorded in Guiana AOE iRobbtns cl a). 2005) The following rasa were not considered in the analysis. ( 1 1 Topaat petta pamprrpur. considered as a synonym with f p. smam&hiD hs f Ti *5* Mtlrgupsii barrinyrrr hybrid between t‘htegop$,s enthropiera and /■ mgromocutat a (Zimmer and Isler 2003); (6) Pmra ifsi snrea'i l.cJb , *' Himugh ihc Amazon; (7) Mionectn uteaginru* huuxwrth race M. o. ImuswriU inseparable from nominate (Fiupitrick MU’ The Wilson Journal of Ornithology 124( l):24-30, 2012 GRASSLAND BIRD COMMUNTIY RESPONSE TO LARGE WILDFIRES ANTHONY J. ROBERTS,1*67 CLINT W. BOAL,2 DAVID B. WESTER,34 SANDRA RIDEOUT-HANZAK.3 AND HEATHER A. WHITLAW5 ABSTRACT.- We studied breeding season communities of grassland birds on short-grass and mixed-grass prairie sites during the second and third breeding seasons following two large wildfires in March 2006 in the Texas panhandle. USA Associated' wlihTfr? 77^' aV,a" CommuQi,y imposition following the tires due to species-specific shifts s3Bmss t n Z , T\VCfm,0n Preferences- SPCC>(* ,hat Prcfer sparse vegetation and bare ground on Meadowlik “ ,° h remophila alpestris ), benefited from wildfires, while others, such as Western srecies'sDeci fic^shif/^*' ”*** deMC Vegeta,ion’ were ^lively impacted. Mixed-grass sites had srimlar’bv "oost h , ’ 7° 7,'"* T*™* ^ f,res; grassland bird communities on burned plots were M'^ng wddfires. Many sbr-” - Fire is a driving ecological process of healthy giassland ecosystems. Following a history of fire suppression, the application of prescribed fire has become a contemporary method to manage grasslands. However, it is unclear how well prescribed fires can replicate the ecological values °f natural fires. For example, studies of grassland bird response to lire have focused on effects of prescribed fire (e.g., Huber and Steuter 1984 Madden et al. 1999. Kirkpatrick et al. 2002. Grant et al. 2010); no published studies have examined the impact of wildfires on avian communities in Great Plains grasslands. Prescribed fires are generally conducted in low wind and high humidity conditions that promote controlled burning. Wildfires may be more intense than presented fires and cover larger spatial scales. Litter is decreased, woody plants are often killed or burned completely, and live plants are killed at high rates (Rideout-Hanzak et al. 2011) This removes structure from the landscape, affecting 'Department of Natural Resources Management. To lech University. Luhbock, TX 79409. USA oC0l0gical Survc>'- Texas Cooperative Fish : S Unit- Te,as T“h ’ Department of Animal. Rangeland, and Wildl A*M n3" KlebCrg Wi,d,ifc Research Institute. Tex ■V S FkhrMta i-,?tSV,,1C' Ki,,8sviUc» 78363. US/ Old Dcpartmenl- 52- Corresponding author; e-mail: tony.roberts@aggicmail.usu.edu 24 (he poorly understood avian community in unknown ways (Smith 2000). There is potential lor increased frequency and size of wildfires in the coming decades due to warmer temperatures and less precipitation, as predicted by current climate change models (North American Bird Initiative 2010). Numerous bird species of concern occur in the short-grass prairies of the southern Great Plains, an assemblage of some of the least ecologically understood of all grassland birds (Askins et al. 2007). The effects of wildfires on these and other species are unknown. Two large wildfires ignited east of Amarillo in the Texas panhandle. USA on 12 March 2006. Together the fires burned 360.000 ha of predom¬ inantly private lands in what is known as the East Amarillo Complex (EAC) wildfires (Zane et al. 2006). Widespread vegetation loss on the mixed- grass and short-grass ecosystems had potentially large negative impacts on numerous species of grassland obligate songbirds that breed in the Southern High Plains. The EAC presented a rare opportunity to examine the effects of large-scale wildfires on grassland bird populations. The objectives of our study were to: (1) examine the changes in avian species densities after a wildfire, and (2) how avian community composition adjusts in the years following the wildfires. METHODS This study was conducted on private ranches in Roberts. Gray, and Donley counties in the Texas panhandle. This area is in the Rolling Plains ecoregion of Texas, a transition zone between short-grass and mixed-grass prairie types in the Southern High Plains. The landscape Roberts et al. • WILDFIRE AND GRASSLAND BIRDS 25 is characterized by rolling hills leading to flat plains interspersed with ephemeral wetland depressions known as playas (Williams and Welker 1966). Elevation ranges from 420 to >600 m. The climate is characterized by hot summers and cold to mild winters, but temper¬ atures fluctuate extensively within seasons (Wil¬ liams and Welker 1966). The mean summer temperature (May-Jul) for 2007 and 2008 was 22.3 C (U.S. Department of Commerce 2009). Historically the Texas panhandle received an average of 53 cm of precipitation a year (Williams and Welker 1966); the study area received ~61 cm of precipitation annually during the study period. Study Plot Selection. — We selected 20 survey plots for study based on access to private property, bum history', and vegetation community type of either short-grass or mixed-grass prairie (cen¬ tered at 14 S 342427 E, 3928890 N). The latter distinction was based on soil type and dominant vegetation. Burn type, either burned or unburned area, was established by talking with landowners and local officials, and visual examination of any woody vegetation in the area such as prickly pear cactus ( Opuntia spp.), catelaw mimosa {Mimosa aculeaticarpa), or sand sagebrush ( Artemisia fdifolia). We were unable to initiate this study and select sites until the fall following the EAC tires and were unable to examine avian commu¬ nities during the breeding season directly after the fires occurred. The 20 plots were equally distributed among areas identified as short-grass burned, short-grass unburned, mixed-grass burned, and mixed-grass unbumed. Individual plots were at least I km apart and 0.5 km from a known fire boundary, roads, or other vegetation or topographic changes. We analyzed each vegetation type separately to examine within-type differences among five replicates each of burned and unburned plots. All study sites were on private property and we were unable to alter grazing regimes; sites ranged from no grazing to moderate slocking levels. Breeding Season Surveys. — We used fixed- radius point counts to survey avian populations within the 20 plots, May-June, 2007 and 2008. We conducted surveys using a three by three grid of nine 75-m radius point counts for a total survey area of ~|6 ha at each plot. We used three observers throughout the study, all of w'hom had either prior experience with grassland bird identification or were trained prior to surveys. Point-count centers were placed 200 m apart to minimize, risk of recounting individuals (Ralph et al. 1 993). We conducted surveys between 0.5 hrs before and 3 hrs after sunrise. We did not survey points during inclement weather such as rain, or in winds >16 km/hr (Ralph et al. 1993). Observers recorded all birds flushed within 75 in of the point when walking to a point. All birds heard or seen in the 75-m radii within a 7-min window' were recorded and the distance from the observer was estimated (Reynolds et al. 1980). Birds seen while walking between points and birds that flew overhead but not using the area within the radii were not recorded. Data Analysis.— We used both the May and June surveys to create a measure of average abundance and species diversity measures over the breeding season. Average abundance was calculated by averaging the count of each species between the two survey periods. We calculated species diversity using the Shannon diversity index (H'; Shannon and Weaver 1963). Taylor (1978) suggested that diversities calculated over a variety of samples arc normally distributed, but we used the more robust /-test suggested by Hutcheson (1970) to compare H' across burn conditions. Hutcheson's /-test uses the number of individuals to calculate degrees of freedom, often resulting in large numbers. We derived evenness (E) from H' using the ratio of observed to maximum diversity as described by Magurran (1988). We only used singing males from June surveys to calculate density estimates. We were only interested in the number of breeding territories supported in each habitat type in contrast to richness and diversity measures. June surveys were during the height of the breeding season when territories were established, and indicative of the density of breeding pairs. We used actual counts of singing males rather than estimates of density derived using detection probabilities. Species-specific estimates of detection probabili¬ ties introduce additional sources of variability (Johnson 2008), and we believed detection probability in a grassland landscape is close to one and similar across bum conditions, especially given our survey was based on territorial males that are announcing their presence (Thompson et al. 2009). Density estimates are an average of the five plots in each treatment. We compared densities among years and burn treatments using a t- test with an alpha level of 0.05. Our interest 26 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 1 ABLE 1 . Avian community measures for breeding season birds among burned and unburned plots associaled with ihe East Amarillo, Texas Complex wildfires of 2006. Short- •grass Mixed- grass Burned In = 5) Unburned In = 5) Bunted In = 5) Unbumed 10% of all species detected overall (Table 2). Effect sizes suggest Lark SpaiTow densities were greater (/8 = 1.18, P = 0.14, d = 0.73) on burned (0.35/ha) compared to unbumed (0.14/ha) plots in 2007: differences between years within plot types were not significant (Table 2). Lark Sparrow densities were similar in 2008 on both burned (0.28/ha) and unbumed (0.15/ha) plots. Moderate effect sizes suggests densities were higher on burned plots in 2008 ( d = 0.57). Roberts et al. • WILDFIRE AND GRASSLAND BIRDS 27 TABLE 2. Avian abundance (percent composition) on burned and unburned short-grass plots associated with the East Amarillo, Texas Complex wildfires of 2006. Short-grass 2007 2008 Species Burned (/i = 5) Unbumed (n = 5) Burned (n = 5) Unbumed (n = 5) Totals Northern Bobwhite (Colinus virginianus) 0 0 1 (0.8) 0 1 (0.2) Scaled Quail ( Callipepla squamata) 0 2(1.9) 0 0 2 (0.4) Killdeer (Charadrius vociferous) 1 (0.7) 0 1 (0.8) 1 (0.8) 3 (0.6) Mourning Dove ( Zenoida macroura) 0 0 4(3.1) 5 (4.3) 9(1.8) Common Nighthawk ( Chordeiles minor) X (5.8) 1 (0.9) 7 (5.3) 5 (4.3) 21 (4.3) ScissOr-tailed Flycatcher (Tymnnus jorfieatus) 2(1.4) 3 (2.9) 2(1.5) 4 (3.4) 1 1 (2.2) Homed Lark ( Eremophila alpestris) 31 (22.4) 15 (14.4) 25 (19.1) 15 (12.9) 86 (17.6) Cliff Swallow (Petrochelidon pyrrhonota) 0 1 (0.9) 1 (0.8) 0 2 (0.4) Bam Swallow ( Hirundo rustica) 1 (0.7) 1 (0.9) 2 (1.5) 0 4 (0.8) Dickcissel i Spiro americana ) 2(1.4) 1 (0.9) 0 0 3 (0.6) Cassin's Sparrow ( Peucaea cassinii) 4 (2.9) 2(1.9) 6 (4.6) 6 (5.2) 18 (3.7) Grasshopper Sparrow ( Ammodramus savannarum) 33 (23.9) 33 (31.7) 27 (20.6) 30 (25.9) 124 (25.3) Lark Sparrow ( Cltondestes grarnmacus) 26(18.8) 11 (10.6) 21 (16.0) 12 (10.3) 70 (14.3) Western Meadowlark ( Stumella neglecta) 28 (20.3) 33 (31.7) 26 (19.8) 33 (28.4) 120 (24.5) Brown-headed Cowbird ( Molothrus ater) 2(1.4) 1 (0.9) 6 (4.6) 5 (4.3) 14 (2.9) Common Crackle (Quiscalus quisculd) 0 0 2(1.5) 0 2 (0.4) Totals 138 131 104 89 490 Mixed-grass.— We observed 12 species on mixed-grass plots in 2007 during the breeding season with nine detected on burned and 1 1 on unburned plots (Table 1). We detected 17 species tut mixed-grass plots in 2008. Bam Swallow (Hirundo rustica). Eastern Kingbird (Ty ramus tyrunnus), Killdeer ( Chamdrius vociferous ), and Lesser Prairie-Chicken ( Tym/mmchus pallidicinc- tus) were all observed in low numbers on burned plots, but were not detected on unbumed plots (Table 3). There was no difference in diversity between burned and unburned plots in 2007. However, diversity on unburned mixed-grass plots in 2008 increased (a6 s = 3.45. P < 0.001) from that in 2007 and was higher (r^s = 2,59, P = 0.01) among burned plots (Table I). No change was detected in diversity between years on burned plots. We analyzed Western Meadowlark. Cassin's Sparrow ( Peucaea cassini /), Lark Sparrow, and Grasshopper Sparrow densities on mixed-grass plots. Western Meadowlark densities ranged from 0.3 1 to 0.45 birds/ha; no differences were detected between bum conditions (/8 = 0.39. P = 0.35) or years (tH = 0.17 ,P = 0.43). Estimated densities of Cassin’s Sparrows in 2007 were 0.36/ha on unburned compared to 0.29/ha on burned plots ({h = 0.35, P = 0.36, d 0.3). Densities did nol change significantly on burned plots (/8 = 0.07. P = 0.47, d = 0. 1 ) or unburned plots (/8 = 0.39. P = 0.35, d = —0.3); Cassin’s Sparrows occurred at identical abundance by 2008 on both plot types (0.30/ha). Grasshopper Sparrows occurred at significantly higher (/8 = 2.2, P = 0.031, d = 1.5) densities on unburned (0.45/ha) than burned plots (0.12/ha) in 2007. Densities numerically decreased (/8 = 1,26. P = 0,12, d = -0.9) in 2008 on unburned plots (0.35/ha) and effect size suggest densities increased (/« = 1.17. P = 0.13, d - 0.8) on burned plots (0.32/ha). Lark Sparrow densities were nearly identical on burned (0.23/ ha) and unburned (0.24/ha) plots in 2007. The following year densities on both burn conditions were 0.15/ha; although the numerical decreases were not significant, it appears there was a noticeable effect of year on both burned (f8 = 0.83. P = 0.22. d = 0.53) or unburned (/8 = 0.85, P =0.21, d = 0.55) plots. DISCUSSION Grasshopper Sparrows were the most abundant species among short-grass plots, and were present in consistent numbers regardless of year or burn condition. This species was significantly more abundant on unbumed plots after a wildfire in Montana shrubsteppe (Bock and Bock 1987), suggesting preference for denser vegetation in arid western landscapes. In contrast, lower numbers of Grasshopper Sparrows were detected on burned areas for 2 years follow ing a grassland 28 THE WILSON JOURNAL OF ORNITHOLOGY • Vo/. 124. No. 1. March 2012 TABLE 3. Avian abundance (percent composition) on burned and unbumed mixed-grass plots associated with the East Amarillo. Texas Complex wildfires of 2006. Species Northern Bobwhite Lesser Prairie-Chicken ( Tympanuchus pallidicinctus ) Killdeer Mourning Dove Common Nighthawk Eastern Kingbird ( Tyrcmnus tyrannus ) Scissor-tuiled Flycatcher Horned Lark Cliff Swallow Barn Swallow Blue Grosbeak (Passerina caerulea) Dickcissel Cassin’s Sparrow Grasshopper Sparrow Lark Sparrow Western Meadowlark Eastern Meadowlark ( Sturnella magna) Brown-headed Cowbird Bullock's Oriole ( Icterus hullockii ) Totals Mixed-grass 2007 Burned (n = 5) Unbumed (n = 5) Burned (n = 5) 1 (LI) 0 0 0 4 (4.5) 0 3 (3.4) 0 0 0 0 3 (3.4) 22 (24.7) 9 (10.1) 18 (20.2) 25 (28.1) 0 4 (4.5) 0 89 2 (1.5) 0 0 3 (2.3) 7 (5.3) 0 0 0 0 5 (3.8) I (0.8) 3 (2.3) 28 (21.1) 35 (26.3) 19 (14.3) 29 (21.8) 0 I (0.8) 0 120 2(1.7) 1 (0.8) 2(1.7) 2 (1.7) 12 (10.0) I (0.8) 2(1.7) 0 0 1 (0.8) 0 0 31 (25.8) 23 (19.2) 12 (10.0) 29 (25.2) 0 2 (1.7) 0 133 2008 Unburned (« = 5) Totals 6 (4.3) 11(2.3) 0 1 (0.2) 0 2 (0.4) 16 (11.4) 21 (4.4) 7 (5.0) 30 (6.2) 0 1 (0.2) 3 (2.1) 8(1.7) 1 (0.7) 1 (0.2) 3 (2.1) 3 (0.6) 0 6 (1.2) 0 1 (0.2) 0 6 (1.2) 27 (19.3) 108 (22.4) 22 (15.7) 89(18.5) 12 (8.6) 61 (12.7) 30 (21.4) 113(23.4) 3 (2.1) 3 (0.6) 8 (5.7) 15 (3.1) 2(1.4) 2 (0.4) 140 482 wildfire in Arizona (Bock and Bock 1992). High effect sizes indicate the higher densities of Western Meadowlarks on burned plots in 2007 may have been biologically relevant, but were not statistically significant in our study. The above¬ ground plant biomass among burned and un¬ burned plots was similar bv 2008 (Rideout- Hanzak et al. 2011), likely providing similar nesting sites and foraging opportunities across burn conditions. Horned Larks prefer areas with a high percentage ol bare ground across their range (Beason 1995). Consistent with this general habitat association. Horned Larks decreased in density as burned areas were revegetated and more litter formed. The composition of avian communities difft on mixed-grass plots in 2007 despite siir diversity and evenness measures across t conditions. Abundances of Cassia's and Gr; hopper sparrows were higher on unburned t burned plots in mixed-grass areas. Howe’ ensities ol these species on burned plots w similar to unburned plots by 2008. Lark Spam- on both short- and mixed-grass plots decreased burned areas from 2007 to 2008. Long-term Li Sparrow abundance decreased along with i decrease in woody vegetation after a fire in sagebrush ( Artemisia spp.) grasslands in Wash¬ ington State (Eamst et al. 2009). This suggests the avian community may have returned to densities similar to unbumed areas by the third breeding season alter the wildfires. A homogeneous landscape with plant biomass and structure similar to unburned areas 3 years after the wildfires ( Rideout-Hanzak et al. 2011) may have promoted similar avian communities across the landscape. Historically, lire was a major ecological factor in both short- and mixed-grass ecosystems until intense livestock grazing and fire suppression altered vegetation and fuels so fires could not burn with historic frequency or intensity (Wright and Bailey 1982). Some areas of the Great Plains have seen lire return to the landscape in the form of prescribed fire. The EAC wildfires occurred during high winds and low humidity, different conditions than proscribed for prescribed fires. Prescribed tires appear to have similar influences on the avian community as the EAC wildfires despite ditterences in intensity and environmental conditions during the wildfire. Many ol the responses measured in our study of wildfire were similar to those observed following Roberts et al. • WILDFIRE AND GRASSLAND BIRDS 29 prescribed fires. Wintering Cassin’s Sparrows in Arizona responded negatively to burning after a prescribed fire, but Grasshopper Sparrows showed no response (Gordon 2000). Abundance of Grasshopper and Cassia's sparrows increased after a prescribed fire in Texas mesquite (Prosopis spp.) savanna (Lee 2006). Prescribed fire in North Dakota mixed-grass prairie decreased populations of most species during the breeding season following the fires (Grant el al. 2010); however. 3 years after the prescribed fires, avian popula¬ tions had increased and stabilized, recovering in a similar time span as in our study. Many species, including Grasshopper Sparrows and Western Meadowlarks, were positively correlated with use of prescribed fire in mixed-grass prairies (Madden et al. 1999). Western Meadowlarks in prairie Canada declined in abundance during the breeding season following prescribed fire, but had comparable densities on burned and unburned areas 3 years post-fire (Pylypec 1991). We found an apparent temporary shift in avian community composition following wildfires due to species-specific shifts associated with life- history traits and vegetation preferences. The avian community appeared to be similar to that on unburned plots of similar grass types 3 years following the wildfires. This was consistent with vegetation recovery (Rideout-Hanzak et al. 201 1 ). A homogeneous landscape in grasslands decreases the diversity of grassland birds (Fuhlendorf et al. 2006) and the grassland bird community reaches peak densities with increased periodic disturbance in short-grass and mixed-grass landscapes. Two of the most common species detected on short-grass plots. Grasshopper Sparrow and Horned Lark, are among 20 common North American birds expe¬ riencing the steepest population declines (Butcher and Niven 2007). The area burned by the EAC wildfires may not only provide important habitat for continued persistence of species of concern, but fire may be an integral component of habitat health for the avian community. Persistence of a diverse and abundant avian community is dependent on periodic disturbances such as wildfire or prescribed fire, grazing, and drought to provide patches of habitat in varying stages of growth after disturbance (Fuhlendorf and Engel 2001). Prescribed fire has been used to mimic wildfire effects and reduce wildfire potential (Pattison 1998). Rideout-Hanzak et al. (2011) suggest the EAC fires may not have created drastically different conditions than a prescribed fire in this ecosystem; this was corroborated by the avian community response. The wildfires may have been ecologically bene¬ ficial in providing similar services to plants and soil as historic fire regimes on the Southern High Plains. The combination of varying grazing regimes and periodic prescribed fire in the Texas panhandle would facilitate development of a mosaic of grassland patches in varying stages of recovery from disturbance, and offer a wide variety of niches for grassland birds. ACKNOWLEDGMENTS We thank the many landowners who allowed access to their properties for this research This study was funded by the Natural Resources Conservation Serv ice. Texas Parks and Wildlife Department, and Texas Tech University. LITERATURE CITED Askins, R. A.. F. Chavez- Ramirez, B. G Dale, C. A. Hass, J. R. Hfrkekt. F. L. Knopf, and P. D. Vickery. 2007. Conservation of grassland birds in North America: understanding ecological processes in different regions, Ornithological Monographs 64:1^16. BEASON, R. C. 1995. Horned Lark ( EremophUa alpestris). The birds of North America. Number 195. Bock. C. E. and J. H. Bock. 1987. Avian habitat occupancy following fire in a Montana shrubsteppe. Prairie Naturalist 19:153-158. Bock, C. E. and J. II Bock. 1992. Response of birds to wildfire in native versus exotic Arizona grassland. Southwestern Naturalist 37:73-81. Butcher. G. and D. K. Niven, 2007. Combining data from the Christmas Bird Count and the Breeding Bird Survey to determine the continental status and trends of North American birds. National Audubon Society, New York, USA, COHEN, J. 1988. Statistical power analysis for the behavioral sciences, Lawrence Earlbaum Associates, Hillsdale. New Jersey, USA. E.arnst, S. L... H. L. Newsome. W. L. La framboise, and N. Laframboi.se. 2009. Avian response to wildfire in interior Columbia Basin shrubsteppe. Condor 111:370-376. Fuhlendorf, S. D. and D. M. Engi.e 2001. Restoring heterogeneity on rangelands: ecosystem management based on evolutionary grazing patterns, BioScience 51:625-632. Fuhlendorf, S. D, W. C. Harrell. D. M. Engle, R. G. Hamilton. C. A. Davis, and D. M. Leslie Jr. 2<)06. Should heterogeneity be the basis for conservation ? Grassland bird response to fire and grazing. Ecological Applications J 6; 1 706- 17)6. Grant. T. a.. E. M. Madden. T. L. Shaffer, and J S Dockets. 2010. Effects of prescribed lire on vegeta- tion and passerine birds in nonhem mixed-grass prairie. Journal of Wildlife Management 74:1841- 30 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 Gordon, C. E. 2000. Fire and cattle grazing on wintering sparrows in Arizona grasslands. Journal of Range Management 53:384—389. Huber. G. £. and A. A. Steuter. 1984. Vegetation profile and grassland bird response to spring burning. Prairie Naturalist 16:55-61. Hutcheson. K. 1970. A test for comparing diversities based on the Shannon formula. Journal of Theoretical Biology 29:151-154. JOHNSON, D. H 2008. In defense of indices: the case of bird surveys. Journal of Wildlife Management 72- 857-868. Kirkpatrick. C„ S. Destefano, R. W. Mannan, and J, Lloyd 2002. Trends in abundance of grassland birds following a spring prescribed burn in southern Arizona. Southwestern Naturalist 47: 282-292. Lee, S. L. 2006. Post-fire successional effect on breeding grassland birds in mesquite savanna habitats of the Texas rolling plains. Thesis. Texas A&M University, College Station, USA. Madden, E. M.. A. J, Hansen, and R. K. Murphy. 1999. Influence of prescribed fire history on habitat and abundance of passerine birds in northern mixed-grass prairie, Canadian Field-Naturalist 1 13:627-640. Magurran, A. E. 1988. Ecological diversity and its measurement. Princeton University Press, Princeton, New Jersey, USA, North American Bird Conservation Initiative. 2010 The slate of the birds 20 |fl report on climate change,” Umtcd Stales of America. U.S. Department of the Interior. District of Columbia. Washington, D. C.. USA. Pattison, M. 1998. Fighting lire with fire: a policy to improve resource management and reduce risk. Renewable Resources Journal 16:13-17. Pylypec, B. 1991, Impacts of fire on bird populations in a fescue prairie. Canadian Field-Naturalist 105:346-349. Ralph, C. J., j. r. Sauer, and S. Droege. 1993. Monitoring bird populations by point counts. USDA. Forest Service, General Technical Report PSW-GTR- 149, Albany. California, USA Reynolds, R. T.. j. M. Scott, and R. a. Nussbaum. 1980. A variable circular-plot method for estimating bird numbers. Condor 82:309-313. Rideou i-Hanzak. s., d. b. Wester, c. M. Britton, and H- Wunt-AW. 201 I. Biomass not linked to perennial grass mortality following severe wildfire in the Southern High Plains. Rangeland Ecology and Man¬ agement 64:47-55. Shannon, C. E. and W. Weaver. 1963. The mathematical theory of communication. University of Illinois Press. Urbana. USA. Smith, J. K. (Editor). 2000. Wildland fire in ecosystems: effects of fire on fauna. USDA, Forest Senile. General Technical Report RMRS-GTR-42. Rocky Mountain Research Station, Ogden. Utah, USA. I aylor. L. R, 1978. Bates, Williams, Hutchinson-a variety o! diversities. In Diversity of insect faunas: 9' Symposium of the Royal Entomological Society (L. A. Mound and N. Warloff. Editors). Blackwell Publishing. Oxford. United Kingdom. Thompson, T. R„ c. w, Boal. and D. Lucia. 2009. Grassland bird associations with introduced and native grass Conservation Reserve Program fields in tbe Southern High Plains. Western North American Naturalist 69:481-490 U.S. Department op Commerce. 2009. National Oceanic and Atmospheric Administration. Silver Spring. Mary¬ land. USA. http://www7.ncdc.noaa.gov/IPS/cd/cd.html Williams, I C. AND A. J. W'elkek. 1 966. Soil survey: Cray County. Texas. USDA. Soil Conservation Service. Washington. D.C.. USA. Wright, H. A. and a. W. Bailey. 1982. Fire ecology. United States and southern Canada. John Wiley and Sons. New York, USA. Zank. D„ .1. Henry, C. Lindley. p. W. Pendergrass. D Galloway. T. Spencer, and M. Stanford. 2006. Surveillance ol mortality during the Texas panhandle wildfires (March 2006). Final Report to Wildland Fire Lessons Learned Center, Tucson. Arizona. USA. www.wildfirelessons.net The Wilson Journal of Ornithology 1 24( 1 ):3 1 —39, 2012 ARTHROPOD ABUNDANCE AND SEASONAL BIRD USE OF BOTTOMLAND FOREST HARVEST GAPS CHRISTOPHER E. MOORMAN,1 4 LIESSA T. BOWEN,' JOHN C. KILGO,2 JAMES L. HANULA/ SCOTT HORN,' AND MICHAEL D. ULYSHEN3 ABSTRACT. — We investigated the influence of aitliropod abundance and vegetation structure on shifts in avian use of canopy gap, gap edge, and surrounding forest understory in a bottomland hardwood forest in the Upper Coastal Plain of South Carolina. We compared captures of foliage-gleaning birds among locations during four periods (spring migration, breeding, post-breeding, and fall migration). Foliage arthropod densities were greatest in the forest understory in all four seasons, but understory vegetation density was greatest in gaps. Foliage-gleaning bird abundance was positively associated with foliage-dwelling arthropods during the breeding (F 18.5, F < 0.001) and post-breeding periods (F - 9.4, P = 0.004). and negatively associated with foliage-dwelling arthropods during fall migration (F = 5.4, P = 0.03). Relationships between birds and arthropods were inconsistent, but the arthropod prey base seemed to he least important during migratory periods. Conversely, bird captures were positively correlated with understory vegetation density during all four periods IF < 0.001 ). Our study suggests high bird abundance associated with canopy gaps during the non-breeding period resulted less from high arthropod food resource availability than from complex underston and midstory vegetation structure. Received 25 January 2011. Accepted $ August 2011. Many bird species, including those of early- successional habitats and those of small tree-fall gaps within mature forest, select disturbed habitats during some portion of the year (Hunter et al. 2001). Several studies have documented greater bird abundance in forest canopy gaps created by natural treefalls (Willson et al. 1982, Blake and Hoppes 1986. Martin and Karr 1986) or group-selection harvest (Kilgo et al. 1999. Moorman and Guynn 2001) than in the mature forest surrounding gaps. Some mature-forest breeders shift into more densely vegetated habitats between breeding and post-breeding periods (Anders el al. 1998: Vega Rivera et al. 1998, 2003; Pagen et al. 2000; Viiz and Rodewald 2006). Birds use a variety of forested habitats during migratory periods (Petit 2000. Rodewald and Brittingham 2002), but mature-forest edges and early-succession habitats may experience relatively greater use (Rodewald and Brittingham 2004). Reasons for greater use of disturbed habitats by birds during certain periods remain speculative, but abundant food and protec¬ tion from predators have been proposed (Marshall et al. 2003). Fisheries. Wildlife, and Conservation Biology Program. Department of Forestry and Environmental Resources. North Carolina State University. Campus Box 7646, Raleigh. NC 27695. USA. L'SDA, Forest Service Southern Research Station, P O. Box 700. New Ellcnlon, SC 29809. USA. USDA. Forest Service, 320 Green Street. Athens, GA 30602, USA. 'Corresponding author; e-mail: chris_moorman@ncsu.edu Arthropod populations also are influenced by season and habitat type (Johnson and Sherry 2001, Greenberg and Forrest 2003) as well as canopy gap size (Shurc and Phillips 1991). It should be advantageous for birds to choose sites with the greatest resource availability (Martin and Kan- 1986), and greater invertebrate biomass has been positively correlated to bird abundance (Blake and Hoppes 1986. Holmes el al. 1986), daily nest survival rates, growth rales of nestlings (Duguay et al. 2000). and liming of warbler migration (Graber and Graber 1983). Studies of experimen¬ tal prey removal have not linked decreased prey abundance with negative consequences for the local bird community (Nagy and Smith 1997, Marshall et al. 2002, Champlin ct al. 2009). Bowen et al. (2007) documented seasonal shifts in relative use by birds of canopy gap and forest habitat. They speculated these shifts may be driven by seasonal changes in arthropod abun¬ dance in gaps. Previous studies have not investi¬ gated seasonal shifts in avian habitat use as related to resource availability over multiple periods. Our objectives were to: (I) investigate whether bird use of forest gaps was associated with arthropod abundance or vegetation structure, and (2) ascertain if shifts in relative use of gap and forest understory were related to spatial and temporal variation in arthropod abundance. We predicted positive relationships between avian habitat use and arthropod abundance (i.e.. relative bird use of gap vs. forest underston will shift based on changes in local arthropod abundance ) from spring migration through fall migration. 31 32 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 METHODS Study Area. — We sampled foliage-gleaning birds and foliage-dwelling arthropods within forest canopy gaps, gap edges, and mature bottomland forest understory during 2001 and 2002 at the Savannah River Site, a 78.000-ha National Environmental Research Park owned by the U.S. Department of Energy. The site was a mature stand of bottomland hardwoods, 120 ha in size, in Barnwell County in the Upper Coastal Plain Region of South Carolina. Birds, arthropods, and vegetation structure were surveyed in 12 group-selection gaps harvested in December 1094 and in the mature forest understory adjacent to gaps. Minimum spacing between gap centers was 100 m. and the mean distance between a gap's edge and the edge of its nearest neighbor was 102.7 m (range = 44-230 m). The gaps were in their seventh and eighth growing seasons post- harvest during the study. They were of three sizes (0.13. 0.26. and 0.50 ha) with four replicates of each size. Previous research within this size range in these gaps identified a threshold in response by breeding (Moorman and Guynn 2001) and fall migrant birds (Kilgo et al. 1999). The mature lorest canopy was dominated by laurel oak ( Quercus lauri folia), cherry bark oak (Q. falcuta var. pagodaefolia), sweetgum ( Uquidamhar styr- aciflua ), and loblolly pine (Pinus taeda). The midstory was patchily developed, consisting primarily of red mulberry {Monts rubra), iron- wood < Carp inns caroUniamis). and American holly (Ilex opaca). The understorv contained patches of dwarf palmetto (Saba! ' minor) and switchcane (Arundinaria gigantea ). Vegetation in the gaps varied from I to 8 m in height and was dominated by regenerating trees (primarily sweet- gum, loblolly pine, sycamore [Platanus occiden- tahs). green ash [Fraxinus pennsy/vanica], oaks, and black willow [Salix nigra]) and patches of blackberry (Rubus spp.). dwarf palmetto, and switchcane. Sampling Design.- We surveyed birds and arthropods during four avian activity periods in , °°' and 2002;. sP,ing migration (25 Mar through 15 May), breeding ( 16 May through 30 Jun), post- breeding (1 jul through 31 Aug), and fall migration (I Sep through 18 Oct). These begin- n>ng and endmg dates are estimates of biologi- Ca ly meanuigfU| periods, and each overlaps SStLdith 'he Ull’Cr Ma"y individuaJs initiated breeding on our study area before 16 May, but transient species that bred to the north continued to migrate through South Carolina until mid-May. Similarly, some individuals mi¬ grated from or through our study area before 1 September, but most fall migration occurred after I September. Wc established a sampling transect radiating southward from the center of each gap to investigate bird-arthropod relationships within each period with three bird and arthropod sampling stations along each transect: one in the gap center, one at the southern edge of the gap. and one 50 m into the forest. Vegetation Measurements. — We measured veg¬ etation structure during June 2001 and 2002 along 10-m transects on each side of and parallel to all mist-net stations. 1,5 m from each net. We measured vertical distribution of vegetation mod¬ ified Irom Karr (1971 ) at I -m intervals along each 1 0-m transect (total 20 points). We recorded the number of times vegetation touched a 2-m pole or the height intervals directly above the pole at 12 height intervals (0-0.25, 0.26-0.50. 0.51-075. 0.76-1, 1. 1-1.5. 1.6-2, 2.1-3. 3.1-5.5.1-10. II- 20, 21-25, and 26-30 in). Touches >2 ni high were estimated visually. The percent cover for each height interval was calculated from the percentage of the 20 sampling points with vegetation touches in that interval. We calculated the mean number of pole touches for height intervals <3 m as an index of foliage density for understory vegetation. Arthropod Collection.— Wc sampled foliage¬ dwelling arthropods at each station during each avian activity period in 2001 and 2002. We used foliage clipping (Cooper and Whitmore 1990) to sample foliage-dwelling arthropods on each oi live target plant species groups: (I) white oaks (white oak 1 Quercus alba), swamp chestnut oak [(?• michauxif). overcup oak | Q. lyrata). Durand oak ]Q. durandii J), (2) lobed red oaks (cherry bark oak). (3) u n lobed red oaks (water oak [ Q . m§ro\. laurel oak. willow oak [<9. p hellos]). (4) sweet- gum. and (5) switchcane. This suite of species was selected to represent dominant members of the understory and overstory, as well as species important as avian foraging substrates (Buffington et al. 2000. Kilgo 2005). Each sample consisted of 25 branch tips from each target species group (total sample = 125 branch tips) collected in the vicinity of each sampling station (i.e.. staying within the target habitat type). Each branch-tip dipping was 2.54-15.24 cm in length and usually came from the end of a branch where most leaves Moorman et al. • SEASONAL BIRD USE OF HABITATS AND ARTHROPODS 33 were clustered on the target plant species groups (Cooper and Whitmore 1990). We collected foliage from ground level to about 2.3 m. We placed clippings immediately in plastic bags to avoid evasive movements of arthropods, hut highly mobile arthropods (a group of less interest lor this study) were not as effectively sampled. We did not sample above 2.3 m because we considered it appropriate to sample arthropods only in the same stratum in which we sampled birds (i.e., 3-m mist nets). Samples were placed in a freezer for 24 hrs to kill all arthropods. We then shook the foliage to collect the arthropods, placed them in alcohol, and identified them to Order. Foliage was oven-dried for 48 hrs at 40 C and weighed. Mist Netting.— We placed a single mist net (12 m long x 3 m tall with 30-mm mesh) at each of the three sampling stations at each of the 12 study gaps, Netting was conducted once each week at each station during the spring migration, post -breeding, and fall migration periods, rotating between stations on a regular weekly schedule. Nets were operated once every 2 weeks during the breeding period, because birds remain fairly stationary during this period. Nets were opened at first light and operated for 4-6 hrs, depending on daily weather conditions. Netting was not conducted when wind exceeded 16 km/hr or during steady rainfall. We banded captured birds with a U.S. Geological Survey aluminum leg band. Statistical Analyses. — We assigned birds (Table 1) to the foliage-gleaning guild following Ehrlich et al. (1988) and Hamel (1992). Birds considered winter residents, present only front late fall through early spring, were not included in analyses. We used a linear mixed model (PROC MIXED) tSAS Institute 2000) to conduct analysis of variance (ANOVA) with covariates and interac¬ tions to analyze the effects of net location (gup. edge, forest understory), period, and arthropod abundance on bird captures. We used mean captures of foliage-gleaning birds/ 1 00 net hrs as the dependent variable. We considered net location and period as fixed effects with net location as a split plot factor and period as the repeated measure. Arthropod abundance was a continuous covariate. We included all two-way interactions. We used a linear mixed model to examine the relationship between bird captures and understory (0-3 m) vegetation density with TABLE 1. Foliage-gleaning bird species captured in mist nets at least once during 2001-2002 in South Carolina. USA. Species Scientific name Yellow-billed Cuckoo Coccyzus americanus Ruby-throated Hummingbird Archilochus coluhris White-eyed Vireo Vireo griseus Blue-headed Vireo V. solitarius Red-eyed Vireo V. olivaceus Carolina Chickadee Poecile carolinensis Tufted Titmouse Bueolophus bicolor Blue-gray Gnatcatcher Polioptila caerulea Gray Catbird Dumetella carolinensis Worm-eating Warbler Helmitheros vermivomm Golden-winged Warbler Vermivora chrysoptera Blue-winged Warbler V. cyanoptera Kentucky Warbler Geothlypis formosa Common Yellowthroat G. trie has Hooded Warbler Setophaga citrina American Redstart S. ruticilla Northern Bum la S. americana Magnolia Warhlcr S. magnolia Chestnut-sided Warbler S. pensylvanica Black-throated Blue Warbler S. caerulescens Pine Warbler S. pinus Prairie Warbler S. discolor Yellow-breasted Chat Icteria virens Summer Tanagcr Piranga rubra Northern Cardinal Cardinalis cardinalis vegetation as the covariate. Vegetation was only recorded once each year, so this model did not include a repeated measure. Year and gap size were not significant (P > 0.05) in any models, and these variables were not included in final models. Arthropod captures were standardized by number of arthropods/ 100 g of dry foliage. We modeled bird abundance with abundance of Lepidoptera because previous studies have shown Lepidoptera to be a primary avian food source (Holmes et al. 1986, McMartin et al. 2002). RESULTS The greatest understory vegetation density occurred in gaps (Fig. 1). Gaps had dense understory vegetation with no canopy, whereas forest had relatively open understory and midstory and well-developed canopy. We captured arthropods representing 21 Orders during 2001 and 2002. Total arthropod density ( number of arthropods/ 100 g of foliage} was lower during spring migration than in the other three periods and greater in the forest understory than in gaps and at gap edges (Table 2). Total arthropod 34 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. 1. March 2012 Mean pole touches per understory (<3 m) height interval FIG. I. Seasonal foliage-gleaning bird caplures plotted against mea Caro^naPUSAnderSt0ry hdght ** CaCh "Ct loca,ion in a understory foliage density (mean vegetation pole bottomland forest during 2001-2002 in South density was greater at gap edges than in g; except during the breeding season when densit were greater in gaps (Table 2). The five m. frequently encountered arthropod Orders rep senting at least 150 individuals, were Arane; Coleoptera, Homoptera, Hymenoptera. and Lei doptera. Aramds. hymenopterans, and lepidoptt ans followed the same general pattern as . arthropods combined, but coleopteran densi decreased from spring to fall migration. Lepido tera density was highest in the forest during sprii migration, breeding, and post-breeding perioc but densities were similar among sampli, locations during fall migration. Bird and arthropod relationships were inconsi tent across the four seasons (Table 3). Folia* g eanmg bird abundance was positively associate «th lol.age dwelhng arthropods during the brecc periods ^ 1-2, P = 0.28) (Table 3). Foliage-gleaning birds were positively associated with understory vegeta¬ tion density during all periods (Table 4; Fig. 1). DISCUSSION Seasonal shifts in relative bird use of gaps and forest understory in bottomland hardwood forests W'ere not driven by changes in arthropod avail¬ ability. Bowen et al. (2007) documented a seasonal shift in habitat use for several bird groups at our site with relative bird use of mature forest habitat greatest during the breeding period: they speculated these shifts may correspond to seasonal changes in arthropod abundance among habitats. However, arthropod abundance remained greater in forest understory than in gaps in all seasons, and we did not document an increase in total arthropods or any arthropod Order in die lores! during the breeding season when relative bird use of torest understory was greatest. The highest arthropod densities in gaps occurred during the breeding season, the period when birds least used gaps. Foliage-gleaning birds on our study site, based on crop flushes, consumed Sti number o( arthropods/ 100 g of dry foliage for most abundant arthropod Orders by period and net location in a bottomland forest during 2001-2002, South Moorman et al. • SEASONAL BIRD USE OF HABITATS AND ARTHROPODS 35 35 CO C3 < -M H i u noonNno;- pi o o' o' - c o +1 +l +1 +1 +1 +1 +1 it) — • — • e-J rn rn rri c- ci O fri rj O ' Ec‘>lc,Sital “‘p °-A and Management Fores‘ Eco,°gy Martin, T. E. and J. R. Karr. 1986. Patch utilization bj migrating birds: resource oriented? Omis Scandari vica 17:165-174. McCarty. J. P„ D. J. Levey. C. H. Greenberg. ai©5 SARGENT. 2002. Spatial and temporal variation in tm use by wildlife in a forested landscape. Forest Ecology and Management 164:277-291. McMartin. B„ I. Belloco. and S. M. Smith. 200! Patterns of consumption and diet differentiation ii* three breeding warbler species during a spnu. budwomi outbreak. Auk 1 19:216-220. Moorf,, F. R.. S. a. Gai hirealx Jk , p. Kerlingef. am J R. SIMONS, 1995. Habitat requirements dunm migration: important link in conservation. Pages I?1 144 in Ecology and management of neotropiui migratory birds (T. E. Martin and D. M. Finch Editors). Oxford University Press, New York. USA. Moorman. C. E. and D. C. Guynn Jr. 2001. Effecb m group-selection opening size on breeding birJ hahiu: use in a bottomland forest. Ecological Applications 1 1:1680-1691. Moorman. C. E., L. T. Bowen, J. C. Kilgo, C. E. Sorenson, J. L. Hanijla. S. Horn, and M. D Ui Y.siii.N, 2007. Seasonal diets of insectivorous bird' using canopy gaps in a bottomland forest, found i" Field Ornithology 78:1-10. Nagy, L. R, and R, T. Holmes. 2005. Food limits annual fecundity of a migratory songbird: an experimental study, Ecology 86:675-681. Nagy. L. R. and K. G. Smith. 1997. Effects of insecticide induced reduction in lepidopteran larvae on reproduc¬ tive success of Hooded Warblers. Auk 114:619-627. Pagen. r. w., f. r. Thompson III, and D. E. Burra, vs. 2000. Breeding and post-breeding habitat use by forest migrant songbirds in the Missouri Ozarks. Condot 102:738-747. Petit. D. R. 2000. Habitat use by landbirds alone Ncari'tie- ncotropical migration routes: implications for conwr vation of stopover habitats. Studies in Avian Biology 20:15-33. REMSEN Jk,, J. V. AND D. A. GOOD. 1996. Misuse of d»u from mist-net captures to assess relative abundance in bird populations. Auk 113:381-398. Rodewald. P, G. ami M. C Brittingham. 2002. Habitat use and behavior of mixed species landbnd flocLs during fall migration. Wilson Bulletin 1 14:87-98. Rodewald. P. G. and M. C. Brittingham. 2004. Stopover habitats of landbirds during fall: use of cdge-doniinatc- and curly-sucoessional forests. Auk 121:1040-1055 SAS INSTITUTE. 2000. SAS. Version 8.1. SAS Institute Inc Cary. NortJi Carolina, USA. Shi re. D. J. and D. L, Phillies. 1991. Patch size of fore*1 openings and arthropod populations. Oecologu 86:325-334. Strong, a. M., T. W. Sherry, .and R. T. Holmes. 2000 Bird predation on herbivorous insects: indirect effect' on sugar maple saplings. Oecologia 125:370-379. Ui v shen, M. D. 2005. The response of beetles (Coleoptcra to group selection harvesting in a southeastern bottomland hardwood forest. Thesis. University of Georgia. Athens, USA. Moorman el al. • SEASONAL BIRD USE OF HABITATS AND ARTHROPODS 39 Ulyshen. M. D.. J. L. Hanula, S. Horn, J. C. Kilgo, and C. E. Moorman. 2004. Spatial and temporal patterns nf beetles associated with coarse woody debris in managed bottomland hardwood forests, l ores! Ecolo¬ gy and Management 199:259-272. Ulyshen, M. D.. J. L. Hanula. S. Horn. J. C Kilgo. and C. E. Moorman. 2005. Herbivorous insect response to group selection cutting in a southeastern bottomland hardwood forest. Environmental Entomology 34:395-402. Ul> shen. M. D.. J. L. Hanula. S. Horn. J. C. Kilgo, and C. E. Moorman. 2006. The response of ground beetles (Coleoptera: Carabidae) to selection cutting in a South Carolina bottomland hardwood forest. Biodiversity and Conservation 15:261-274. Vega Rivera. J. H.. J. H. Rappole. W. J. McShea. and C. A. Haas. 1998. Wood Thrush postfledging movements and habitat use in northern Virginia. Condor 100:69- 78. Vega Rivi-ka, J. H„ W. J. McSiiea. and J. H. Rappole. 2003, Comparison of breeding and postbreeding move¬ ments and habitat requirements for the Scarlet Tanager (Piranha oHvaeea) in Virginia. Auk 120:632-644. V my. A C. AND A. r>. RODEWALD. 2006. Can regenerating dearcuts benefit mature-forest songbirds'.’ An exami¬ nation of post-breeding ecology. Biological Conserva¬ tion 127:477-486. Willson, M. F„ F.. a. Porter, and R. S. Condit. 1982. Avian frugivorc activity in relation to forest light gaps. Caribbean Journal of Science 18:1-6. Wilson, R. R. and D. .1. Twhdt. 2003. Spring bird migration in Mississippi Alluvial Valley forests. American Midland Naturalist 149:163-175. The Wilson Journal of Ornithology 124(l):40-50. 2012 SEASONAL VARIATION IN SHOREBIRD ABUNDANCE IN THE STATE OF RIO GRANDE DO SUL, SOUTHERN BRAZIL ANGELO L. SCHERER1-2 AND MARIA V. PETRY' ABSTRACT — We describe the frequency of occurrence and seasonal variations of shorebirds (Charadriidae and Scolopacidae) along a 120-km transect of beach between Balnefirio Pinhal and Mostardas north of Lagoa do Peixe National Park. Rio Grande do Sul State over a 2-vear period (Oct 2007 to Sep 2009). A total of 96.889 shorebirds was recorded. The greatest abundance occurred between October and April and the lowest occurred between May and September. The most abundant of the 17 species recorded were Sanderlings (Calidris alba). White-rumped Sandpipers tC fuscicollis ) . and Red Knots (C. camttus). The least abundant were Senupalmated Sandpipers (C. pusilta). Rufous-chested Ploxers ( Charadrius mode slits), and Hudsonian Godwits ( Li most/ haemastiea). Fourteen species were migrants from the Northern Hemisphere, one was a migrant from the Southern Hemisphere, and two were residents. Nine species were recorded regularly, two were recorded sporadically, and six were recorded occasionally. Six Nearctic species were recorded in June and July most likely indicating the presence of non-breeding immatures. The beaches of Rio Grande do Sul are important migration stopover and wintering sites for many shorebirds in southern Brazil and should be a focu> of conservation efforts, especially given the increasing development pressure that threatens these areas. Received 10 February 2011. Accepted 25 October 2011. Brazil is visited by thousands of birds that migrate seasonally from the Northern lo Southern Hemisphere and vice-versa (Morrison and Ross 1989, Chesser 1994). Those that come from the north prior to the boreal winter (Antas 1994) arrive in Brazil seeking wintering sites rich in food resources (Telino-Junior et al. 2003). Shore- birds from the Northern Hemisphere occur in Brazil during the austral summer and those from the Southern Hemisphere occur in the country during the austral winter. Immature individuals of certain shorebirds occur throughout nearly the entire year, as they are not yet capable of breeding and return to breeding areas only when they are sufficiently mature to begin nesting (Sick 1979. Azevedo-Junior and Larrazabal 1994. Azevedo- Junior et al. 2001a. b). The sites at which shorebirds stop during migrations are of considerable importance for conservation. Lagoa do Peixe National Park and nearby beaches on the coast of the State of Rio Grande do Sul in southern Brazil attract large concentrations of migratory shorebirds (Lara- Resende and Leeuwenberg 1987. Morrison and Ross 1989. Vooren and Chiaradia 1990. Belton 2000). The availability of food resources at these sites oilers the birds the opportunity to gain body Graduate Program in Biology. Laboratory of Ormtl ogy and Marine Animals. Center for Health Scienc Va'1' d° Ri,> dos Si™s- Avenida Unisin . ‘ °‘ ^0'-2-°00. P. o. Box 275. Room 2D223E. 5 Leopoldo. Rio Grande do Sul, Brazil Corresponding author; e-mail: alscherer@pop.com.bi mass and acquire adequate energy for molting and return to breeding areas (Azevedo Junior et al. 2001a, b; Baker et al. 2004, Fedrizzi et al. 2004). Shorebird populations fluctuate during (he breeding and wintering periods in number of individuals and migratory species using the beaches of southern Brazil (Lara-Resende and Leeuwenberg 1987, Barbieri et al. 2003. Telino- Junior et al. 2003. Barbieri and Mendon^a 2005) The greatest abundance occurs from September to April along the Brazilian shoreline with slight temporal variations (Barbieri 2007. Barbieri and Hvenegaard 2008. Barbieri and Paes 2008). In southern Brazil. Lara-Resende and Leeuwenberg 1 19X7) found the highest abundance of shorebirds from November to April in Lagoa do Peixe National Park (Ramsar site) (31 21’ S; 05 1 03’ W). I he beaches from Balneario Pinhal to Mostardas. which are north of this conservation unit, offer adequate stopover, feeding, and resting sites for migratory shorebirds and are important locations for their conservation (A. L. Scherer, pers. obs.). The coast of Rio Grande do Sul is considered a key area for shorebird conserv ation in the Western Hemisphere (Serrano 2008). Our objectives were to: ( I ) record the occurrence and seasonal variations ot shorebirds (Charadriidae and Scolo¬ pacidae) along a 120-km transect of beach north of Lagoa do Peixe National Park, and (2) document the frequency of occurrence of species of shorebirds over a 2-year period. 40 Scherer and Petry • SEASONAL VARIATION IN SHOREBIRD ABUNDANCE 41 0 15 30 6Q Mostardas Lagoa do Peixe National Park 51 WW Balneario Pinhal -f PARAGUAY BRAZIL L-'v 1 \ S - f ( J ARGENTINA J* / RO GRANDE DO SUL TOMES BAI MiAMpmUAI 1 ■ URUGUAY -\ •y . \. \ « _ < - V-. j/ Legend Lakes Agriculture Urban area Dunes and Fields Forest -31 ws — i - J9-OCTW TkJ. I, Transect (120 km) of study area between municipalities of Balnedrio Pinhal (30J 14' 57' S. 050 13 48.4 W) and Mostardas (31 10' 52" S. 050 50’ 03” W) on the coast of Rio Grande do Sul. southern Brazil. METHODS S'/u/v Area.— We studied shorebirds along the coast °I the State of Rio Grande do Sul (southern Brazil) between the municipalities of Mostardas 1,1 the south (31 10' 52" S. 050 50' 03" W) and Balneario Pinhal in the north (30 14' 51" S. 050 1 48.4” W), totaling 120 km of beach (Fig. I). The area is characterized by large sandy beaches dlul Stines with a series of coastal lakes beyond !i>e dunes (Belton 2000). Human occupation along 'his region has altered the landscape, which is US«J (or tourism and recreation. Fifteen kilome- ,ers °f shoreline in Balneario Pinhal are highly urbanized and the entire region is used intensively lor irrigated crops and tree plantations. The climate of the region is wet subtropical VVl|h a mean temperature of 17 C and annual 'ainfall of 1 .200 mm. The width of the 120-km s,reteh of beach ranges from 50 to 1 20 m. at times reaching 200 m. The beach has a low slope and 'he wash zone is broad, generally -10 m with a high density of invertebrates. The dominant species are bivalve mollusks including yellow clams ( Mesodesma mcictroides) and wedge clams ( Donax hanleyamis), crustaceans such as mole crabs (Enterita brasiliensis) and cirolanid isopods ( Excirolana armata), and polychaetes ( Euzonus furciferus and Spin gaucha) (Gianuca 1983). Surveys.— Monthly surveys (n = 24) were conducted between October 2007 and September 2009 along the 120 km of beach in an automobile traveling from north to south at a maximum speed of 20 km/hr between 0700 and 1700 hrs. Direct counts were made of the individuals of each spe¬ cies (Bibby et al. 2000). Surveys were conducted by two observers on randomly chosen sunny days. Shorebird abundance was recorded at a distance (50 to 100 m) to keep birds from Hushing. The first observer recorded the birds from the edge of the water to the middle portion of the beach and the second observer recorded the birds from this portion to the dunes, taking care to avoid recounts. Data were recorded on a portable voice recorder and field spreadsheets. Birds were identified with 42 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 TABLE 1. Occurrence and total abundance of shorebirds (in decreasing order) from October 2007 to September 2000 on the coast of Rio Grande do Sul. southern Brazil. Status: seasonal visitor from the Northern Hemisphere (VN). seasonal visitor from the Southern Hemisphere (VS), and Resident ( R ). Species Slums Abundance * Sanderling Calidris alba VN 78.247 80.8 White-rumped Sandpiper C. fiuscicollis VN 7.588 7.8 Red Knot C. canutus VN 5,103 5.3 Collared Plover Charadrius collaris R 2.042 2.1 Lesser Yellowlegs Tringa flavipes VN 1.517 1.6 Grey Plover Pluvialis squatarola VN 822 0.8 Semipalmated Plover Charadrius semipalmatus VN 547 0.6 Southern Lapwing Vanellus cliilensis R 481 0.5 American Golden Plover Pluvialis dominica VN 218 0.2 Ruddy Turnstone Arenaria inierpres VN 127 0.1 Greater Yellowlegs Tringa rnelanoleuca VN 95 0.1 Solitary Sandpiper T. solitaria VN 36 0.04 Buff-breasted Sandpiper Tryngites subrupcoUis VN 35 0.04 Whimbrel Numenius phaeopus VN 17 0.02 Semipalmated Sandpiper Calidris pusilla VN 7 0.01 Rufous-chested Plover Charadrius modestus VS 6 0.01 Hudsonian Godwit Limosa haeinastica VN 1 0.001 Totals 96.889 100 the aid ol 10 X 50 binoculars and use of field guides (Crossley et al. 2006, Mata et al. 2006). Flocks of shorebirds with <100 individuals were considered small, those with between 100 and 1,000 individuals were considered medium-sized, and those with >1,000 individuals were consid¬ ered large. The seasons are in accordance with the Southern Hemisphere: austral summer (Oct to Mar) and austral winter (Apr to Sep). The frequency of occurrence (C) of each species was calculated using the equation C = 1J x 1 00/A' in which P is the number of counts containing the species, and N is the total number of counts throughout the study period ( n = 24). Classifica¬ tions recorded were: regular (present in >50% of counts), sporadic (present between 25 and 50% of counts), occasional (present <25% of counts), and absent (not present in counts) (Dajoz 1983). RESULTS Seventeen species of shorebirds were recorded, six species of Charadriidae and 1 1 Scolopacidae, totaling 96.889 individuals. Sanderlings (Calidw alba). White-rumped Sandpipers (C. fuscicoUh). and Red Knots (C. canutus) were the most abundant species, accounting for 93.9% of the overall abundance. Sanderlings accounted for 80.8% of the overall abundance. Collared Plover ( Charadrius col laris) and Lesser Yellowlegs (Tringa flavipes) accounted for >1% of the overall abundance, whereas the remaining 12 species accounted for <1% (Table 1 j. The greatest abundance of shorebirds occurred between October (arrival at stopover or wintering areas) and April (return to breeding areas), and the lowest occurred between May and September (Fig. 2). Nine species were recorded regularly over a 2-year period (2007 to 2009) (all but the Red Knot were also regular when years were analyzed separately/ two were sporadic, and six were occasional (Table 2). The Hudsotiian Godwit (Limosa haemastica ) was absent throughout the 2007/2008 period and only one individual was recorded in January 2009. The Rufous-chested Plover ( Charadrius modestus). a visitor from the Southern Hemisphere, was recorded as occurring occasionally. One individual was recorded in July 2008 and five were recorded in May 2009, Five Semipalmated Sandpipers (Calidris pusilla ) were observed in February 2008 and two in September 2009. Six Buff-breasted Sandpiper^ (Tryngites mbruficollis) were observ ed in November 2007 and 29 (distributed among three sites) were recorded in October 2008. Abundances during monthly surveys varied seasonally (Fig. 3). The frequency of occurrence of the Southern Lapwing ( Vanellus chilensis), a resident species, was regular as it was recorded in all counts. The greatest abundance of this species occurred between April aid July with a peak (n = 81) in July 2008. and was lowest between October and December with only one individual recorded in November 2008. The Amer¬ ican Golden Plover (Pluvialis dominica) was occa¬ sional but present in 4 months with a peak (n = 126) in February 2009; six individuals were recorded in April in the 2007/2(X)8 period. The frequency of occurrence of Grey Plovers (P. squatarola) was Scherer and Petry • SEASONAL VARIATION IN SHOREBIRD ABUNDANCE 43 14000 ■ 12000 ■ 10000 ■ o 8000 ONDJ FMAMJJASONDJ FMAMJ J AS 2007- 2009 FIG. 2. Monthly abundance of shorebirds (Charadriidae and Scolopacidae) from October 2007 to September 2009 on the coast of Rio Grande do Sul. southern Brazil. regular with greatest abundance occurring between October and February (n = 353 in Feb 2009) tuid lowest between April and September, this species was absent in June and July. The frequency of occurrence of Semipalmated Plovers (Charadrius semlpalmutus ) was sporadic '» the 2007/2008 period and regular in 2008/2009. ihe greatest abundance was recorded between February and March 2008, and between April and May 2009 with no records of occurrence between (Mober and January in either year. The frequency of occurrence of Collared Plovers was regular and the species was recorded in all months. The greatest abundance of this species occurred between March and August, and was lowest between September and February. A large part of the increase in abundance beginning in March was due to the presence of young of the year, which were observed foraging in the w'ash zone and near the drainage sandbars of lakes. Abun¬ dance was considerably higher in May, June, and August 2008 than during the same period in 2009. TABLE 2. Frequency of occurrence (C) of shorebirds from October 2007 to September 2009 Orande do Sul, southern Brazil. Oct 2007 to Sep 2008 Oct 2008 to Sep 2009 Specie* c Constancy C Constancy c Southern Lapwing 1 00 Regular 100 Regular 100 American Golden Plover 8 Occasional 33 Sporadic 21 Grey Plover 58 Constant 83 Regular 71 Semipalmated Plover 33 Sporadic 50 Regular 42 Collared Plover 100 Regular UK) Regular 100 Rufous-chested Plover 8 Occasional 8 Occasional 8 Uudsonian Godwii 0 Absent 8 Occasional 8 'A'himbrel 0 Absent 33 Sporadic 17 Solitary Sandpiper 33 Sporadic 17 Occasional 25 Greater Yellowlegs 50 Regular 67 Regular 58 ^csser Yellowlegs 83 Regular 92 Regular 88 Ruddy Turnstone 67 Regular 58 Regular 63 Red Knot 67 Regular 33 Sporadic 58 Sanderling 100 Regular 100 Regular 100 Semipalmated Sandpiper 8 Occasional 8 Occasional 8 vvhiic-rumped Sandpiper 83 Regular 67 Regular 75 Ruff-breasted Sandpiper 8 Occasional 8 Occasional 8 2007 to 2000 Constancy Occasional Occasional 44 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. I. March 2012 FIG. do Sul. 2007-2009 2007-2009 2007-2009 20 18 16 • 14 Jl2 fio < 8 6 4 2 0 • Solitary Sandpiper J ONDJFMAMJJAsONDJFMAMjJAS 2007-2009 ^ 2007-2009 . southern Brazil. *" °* shorehirds sPecies from October 2007 to September 2009 on the coast of Rio Grande Scherer and Petry • SEASONAL VARIATION IN SHORE BIRD ABUNDANCE 45 2007-2009 80 70 60 ti v =50 v =40 £ <30 20 10 0 Ruddy Turnstone 111 ONDJ FMAMJJASONDJFMAMJJAS 2007-2009 2007-2009 F,G- 3. Continued. The frequency of occurrence of Whirabrels l ‘'Wnu« phaeopus) was occasional as the species was absent in 2007/2008 and sporadic in -008/2009. The greatest number of individuals ■v*» recorded in November 2008 (n = 6) and May -009 i/i = 9) only one individual was observed in«»ch of two other months. Solitary Sandpipers 'Thnga solitaria) occurred sporadically in 2007/ 4*08 and occasionally in 2008/2009. This spe- c*es had no clear tendency of occurrence, but was 2007 - 2009 recorded in two consecutive months (Nov and Dec 2007) with greatest abundance (n = 18) in November 2007. The frequency of occurrence of Greater Yel- lowlegs (T. melanoleuca) was regular over both years. The greatest abundance occurred between May and September with a peak in July 2008 tn = 16) and in September 2009 (n = 16). This species exhibited abrupt variation from one month to the next as observed in May. June, and July 2008 w ith 46 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. 1. March 2012 13, 1, and 16 individuals observed, respectively. December. March, and April were the only months in which there were no records of this species. The frequency of occurrence of Lesser Yellowlegs was regular over both years with greatest abundance between May and October, and lowest between December and April. Abun¬ dance beginning in May 2008 increased the following month and then decreased until Sep¬ tember with a significant increase in October 2008, reaching 270 individuals before rapidly declining in November. This species was absent in October and November 2007. and April 2009. Ruddy Turnstones (Aren aria interpret,) were regularly observed every month of the 2-year period, and had low abundance ranging from one to six individuals. However, 15 individuals were observed in December 2007 and the greatest abundance occurred in April 2009 with 70 in¬ dividuals distributed among three flocks of 30, 15. and 25 individuals spatially separated I and 7 km from one another, respectively. The Red Knot was the ihird most abundant species and was observed regularly in the first year and sporadically in the second year. The greatest abundance occurred between April and September with uneven distribution among the remaining months of the year. Monthly abundance in 2008 ranged from 1,669 individuals in April to 96 individuals in May, and rising in June and July. A considerable number of individuals (/? = 322) was recorded in April 2009. but only four were observed in May, none in June or July, before becoming abundant in September (n = 1.562). This species w'as absent in December and March with only one individual recorded in February 2009, two individuals in October 2007, and two in January 2008. Thirty-eight and 71 Red Knots were observed in June and July 2008, respective¬ ly; some were in nuptial plumage while in flocks with non-breeding plumage birds, which were likely not migrants that year. The Sanderling was the most abundant species with regular frequency of occurrence. The greatest numbers occurred between September and April with an abrupt increase and decline at the be¬ ginning and end ol this period. Peaks (// = 5.943) were recorded at the beginning of the austral summer (Nov 2007) and beginnine (n = 10 757) of theifus>™' winter (Apr 2008); the same pattern in °C'0ber 2I,0H = s-2%' and Apr.! 2009 („ = 9,759). The lowest abundance occurred between May and August, and flocks most likely consisted of non-breeding immature individuals. The lowest numbers were recorded in July (n = 1) and August (// - 3) 2008. The frequency of occurrence of White-ramped Sandpipers was regular and this species was the second most abundant. This species also had two yearly peaks of abundance with largest numbers in November 2007 (n = 1,204) and March 2008 (n = 1,060). and in August 2008 ( n = 561) and February 2009 (n = 1,012). This species was absent in May. June, and July in both years. DISCUSSION The presence of 1 7 species of shorebirds, some highly abundant, confirms the importance of the beaches (Balncario Pinhal to Mostardas) on the coast of Rio Grande do Sul State (southern Brazil) as stopover and wintering sites for migratory shorebirds. Fourteen species of shorebirds record¬ ed in the present study were migratory that breed in the Northern Hemisphere. These shorebirds use the beaches of Rio Grande do Sul as stopover or wintering areas for feeding and resting, mainly between October and April. This period coincides with greater availability of macroinvertebrates along the beaches and banks of coastal lakes ( Lara-Resende and Leeuwenberg 1987). when lower water levels allow greater foraging area. The number of shorebirds was lowest between May and September when most populations migrate to breeding areas in the Northern Hemisphere. However, a small number of repro- ductively immature juveniles (Semipahnated Plo¬ ver. Greater Yellowlegs, Lesser Yellowlegs. Red Knot, Sanderling, and White-rumped Sandpiper) remain on the beach throughout the austral winter, as has also been reported for beaches in northern and northeastern Brazil ( Azevedo-Jtinior and Larra/abal 1994, Azevedo-Junior et al. 2001a. b: Barbicri and Mendon^a 2005. Barbieri and Hvenegaard 2008). 1 he most frequent and abundant species on the beaches between Balncario Pinhal and Mostardas were Sanderlings, White-rumped Sandpipers, and Red Knots. These species have previously been recorded on other beaches of Rio Grande do Sul ( Lara-Resende and Leeuwenberg 1987. Vooren and Chiaradia 1 990). The Rio Grande do Sul beaches ate among the most important wintering areas on the Atlantic Coast of South America for Sanderling1' (Morrison and Ross 1989). The three species are highly abundant in Lagoa do Peixe National Park (Lara-Resende and Leeuwenberg 1987), using the Scherer and Petry • SEASONAL VARIATION IN SHOREBIRD ABUNDANCE 47 area as either a wintering site or stopover point for the part of the population that winters in areas further south, such as Tietra del Fuego (Harrington et al. 1986. Morrison et al. 200), Piersma 2007). Nine species had a regular frequency of occurrence over the 2-vear period, whereas other species occurred occasionally and in low abun¬ dance (Rufous-chested Plover. Hudsonian God- wit. .Semipalmated Sandpiper, and Buff-breasted Sandpiper). The Rufous-chested Plover is an austral winter visitor from Patagonia, where it breeds (Belton 2000). This species occurs in low abundance on the beaches of southern Brazil lLara-Resende and Leeuwenberg 1087, Vooren and Chiaradia 1990, Belton 2000) and Argentina iBIanco et al, 2006). Hudsonian Godwits occur throughout the year in Lagoa do Peixc with highest abundance in November during its migration south (up to 1.300 individuals) and in March prior to migration north (Lara-Resende and Leeuwenberg 1987). Up to 3.000 individuals have been recorded at the site (Harrington et al. 1986. Morrison and Ross 1989). This species is highly sensitive to disturbed environments (Parker et al. 19%) and is rarely seen on beaches with people and vehicles. The Semipalmated Sandpiper occurs with low frequency and abundance in southern Brazil (Lara-Resende and Leeuwenberg 1987, Vooren and Chiaradia 1990. Costa and Sander -(||)8), but is frequent and abundant on beaches of northern and northeastern Brazil, where it winters Morrison and Ross 1989, Azevedo- Junior et al. 20Ulb, Telino-Junior et al. 2003, Barbieri 2007). The Buff-breasted Sandpiper is considered endangered (BirdLife International 201 i ) and 0CCUrs in wet grassland areas and grazed pasturelands near lagoons (Lara-Resende and Leeuwenberg 1987. Lanctol et al. 2002). This species generally migrates through the interior the continent, using the central Amazon/ Pantanal wetland route, where it follows large rivers and wetlands until reaching Paraguay and Argentina (Antas 1984). This explains the low frequency and abundance on the beaches between Balneario Pinhal and Mosturdas. where 11 recorded on only two occasions (Nov 2007 and Oct 2008). The resident Southern Lapwing, which is Cndemic to South America, was among the species *rith regular frequency of occurrence and was observed on all counts. This species is common in grassland areas of Rio Grande do Sul. where it breeds, and Belton (2000) considered its occur¬ rence on the shore as rare. It is observed with greater frequency foraging on the beach near sandbars with drainage of waters from lakes between April and July: lower numbers are observed between October and December when this species is in its breeding season (Belton 2000). The American Golden Plover had a lesser frequency of occurrence and lower abundance than the Grey Plover with greatest abundance occurring between November and March, as previously reported for wintering areas in south¬ ern Brazil (Lara-Resende and Leeuwenberg 1987, Vooren and Chiaradia 1990). This species is found mostly on beaches within the proximity of Lagoa do Peixe (Morrison and Ross 1989) where they encounter habitat with less direct influence from humans and vehicles along the beach. The Grey Plover had low frequency and abundance along the beach throughout the year (Lara- Resende and Leeuwenberg 1987, Vooren and Chiaradia 1990). The frequency of the Semipalmated Plover was uneven with greater abundance in February and March 2008, and April and May 2009. However, this species has been observed in low abundance in southern Brazil (Lara-Resende and Leeuwen¬ berg 1987, Vooren and Chiaradia 1990). It is more abundant on beaches of northeastern Brazil, where most of the population winters (Rodrigues 2000, Telino-Junior et al. 2003, Barbieri 2007. Barbieri and Hvenegaard 2008). The resident Collared Plover, which breeds among the dunes along the Brazilian coast, occurs throughout the year. Little is known about this species. Its greatest abun¬ dance on the beaches between Balneario Pinhal and Mostardas was between March and August. This was during the breeding season, which extends from November to January, when the species is frequently found foraging near dunes where it breeds, as well as throughout (he non¬ breeding season in the austral winter (Lara- Resende and Leeuwenberg 1987, Belton 2000). Juveniles were observed foraging in (he wash zone and drainage sandbars of lakes in every month of the year, but with a considerable increase beginning in March and greater abun¬ dance in the subsequent months. Whimbrel occurred in low frequency and abundance. This species needs conserved envi¬ ronments due to high sensitivity to environmental disturbances (Parker et al. 1996), The largest flocks occur in the northern and northeastern regions of Brazil , and northern portion of South 48 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 America (Morrison and Ross 1989. Azevedo- Junior and Larrazabal 1994. Azevedo-Junior et al. 2004). This species in southern Brazil is only observed foraging in small numbers in Lagoa do Peixe (// < 8) (Lara-Resende and Leeuwenberg 1987) and in the surrounding beaches. Solitary Sandpipers were observed sporadically and in small numbers along the beach, as reported by Costa and Sander (2008) on other beaches in Rio Grande do Sul. Greater Yellowlegs were observed regularly in both years as reported by Lara-Resende and Leeuwenberg (1987) with peaks of abundance in July and September, but was much rarer than Lesser Yellowlegs. Many non-migrating individuals of Greater Yellowlegs that winter in interior areas of the state, according to Belton (2000). concentrate on the beach during the austral winter. Lesser Yellowlegs are recorded in every month of the year in Rio Grande do Sul. appearing rarely during the austral winter and more commonly from September to March (Belton 2000). However, we found the species was more abundant between May and October and was rare between November and April. Those observed from May to July were likely flocks of juveniles that had wintered on the beaches and lakes in the interior of Rio Grande do Sul and then concentrated on the coast. The greatest abundance in October coincides with migration to South America with the species using beaches and lakes as either wintering or stopover areas. Lesser and Greater yellowlegs were observed foraging on the banks of small lakes near dunes and drainage sandbars of these lakes and. at times, in the wash zone. The Ruddy Turnstone is common along the coast of the Province of Buenos Aires (Myers and Myers 1979. Blanco et al. 2006) during the austral summer. We found low abundance of this species on the beaches from Balneario Pinhal to Mostar- das with greater frequencies occurring between April and July, and between September and December. This has also been reported in previous studies (Harrington et al. 1986, Lara-Resende and Leeuwenberg 1987. Voorcn and Chiaradia 1990). This indicates that small groups stop to feed and rest during migrations south and north between wintering and breeding areas in Argentina and the Northern Hemisphere, respectively. The pattern of occurrence of the Red Knot corroborates that observed at other sites in southern Brazil (Lara-Resende and Leeuwenberg 1987, Vooren and Chiaradia 1990). The greater abundance in September indicates arrival of migrants from the Northern Hemisphere moving toward wintering areas in Patagonia and Tierradel Fuego. This species use the lakes and beaches of Rio Grande do Sul as stopover areas for feeding, resting, and molting. A few individuals remain in winter and join flocks that stopover in April on return to breeding areas in the Northern Hemi¬ sphere. The stopover period on the beaches from Balneario Pinhal to Mostardas is short, but of considerable importance to the migratory process between the extremes of the two hemispheres (Piersma 2007). This beach serves as one of the most important stopover sites during the molting process and for gain in body mass (Mormon and Ross 1989, Vooren and Chiaradia 1990). The regular frequency throughout the year and high abundance of Sanderlings confirms the bea¬ ches of Rio Grande do Sul are one of the mow important wintering areas for this species along the Atlantic Coast of South America (Harrington el al. 1986. Lara-Resende and Leeuwenberg 1987, Morrison and Ross 1989, Myers et al. 1990i. Large flocks begin to arrive in September with peak abundance in October and a decrease in the following month as the areas are used as a stopover site during migration farther south to wintering areas. The abundance of Sanderlings remains stable from November to February because of individuals that winter in the area, decreases from February to March, when these same individuals begin to migrate north, and reaches its largest peak in April, when those that wintered to the south (Argentina and Patagonia1 again use the site as a stopover, and declines in May. We found the species in greater densities than at other sites in southern Brazil (Vooren and Chiaradia 1990. Costa and Sander 2008). indicat¬ ing greater availability of food resources on these beaches and lower human impacts. White -rumped Sandpipers occurred in number* and pattern of occurrence similar to that of the Red Knot. This species is reported in Lagoa ck Peixe to be highly abundant between November and January, both on the lake and the beach (Lara- Resende and Leeuwenberg 1987). and part of the population uses the site as stopover and wintering areas (Morrison and Ross 1989). The greater abundance occurs at the beginning of the austr.il summer with arrival of migrants from !h< Northern Hemisphere to winter in Lagoa do Pci'1' and Argentina, where they are abundant through out the austral summer (Myers and Myers 1979. 1 Scherer and Petry • SEASONAL VARIATION IN SHOREBIRD ABUNDANCE 49 Lara-Resende and Leeuwenberg 1987, Morrison and Ross 1989. Blanco et al. 2006). We observed a few individuals on the beaches from Balneario Pinha] to Mostardas during the austral summer with abundance increasing at the beginning of the austral autumn (Mar and Apr), when the popula¬ tion returns from wintering areas on migration to North America. This indicates the importance of these beaches for stopover and feeding sites prior to migration. CONSERVATION IMPLICATIONS Knowledge of stopover and wintering areas is critical to the conservation of migratory shore- birds. as these sites require special care (Lara- Resende and Leeuwenberg 1987, Myers et al. 1990) through an internationally coordinated effort directed al protecting breeding, stopover, and wintering areas (Morrison 1984, Morrison et al. 2004). The beaches from Balneario Pinhal to Mostardas in the Stale of Rio Grande do Sul (southern Brazil) are used by large numbers of shorebirds as stopover and wintering areas, especially Sanderlings, White-rumped Sandpip¬ ers and Red Knots. Our study underscores the importance of these beaches as priority areas for implementation of conservation projects directed at these shorebirds and serves as a basis for future comparisons addressing the impact of environ¬ mental changes on these species (Piersma and Lindsirorn 2004). ACKNOWLEDGMENTS Ihe authors are grateful to the Wildlife Conservation s"ciei.v (WCS) for funding (his study (Contract « 2008005). v L Scherer is grateful to the Brazilian fostering agency 1 'Vinlena^ao de Apcrfciyoamcnto de Pcssoal de Ensino 'jperior (CAPES) for a study grant awarded from the 1 digram a de Supoiie h Pds-Graduayao de InstituiyoCS de Ensino Paniculares (PROSUP). We thank R. (i. de Mourn 11 'r help drawing Figure I. and S. M. A/cvedo-Junior and L H. Oliveira for helpful comments and reviews. LITERATURE CITED p. T. Z. 1984. Migration of Nearclic shorebirds 'Churudriidae and Scolopacidac) in Brazil — llywa>s ‘tnd their different seasonal use. Wader Study Group Bulletin 39:52-56. Astas. p. t. Z 1994. Migration and other movements among the lower Paranft River Valley wetlands. Argentina, anil the south Brazil/Paiitanal wetlands. Birdlife Conservation International 4: IK 1-1 90. AZF.VEDO- j ONIOR, S M. AND M- E. I.ARRA7.ABAL. 1994. Censo de aves lirotcolas na Coroa do Aviao, Pernambuco, Brasil, informayoes dc 1991 a 1992. Revista Nordestina de Zoololgia 1:263-277. Azevedo-Jinior. S. M.. M, M. Dias, and M. E. Larka/ahai.. 2001a. Plumagens e mudas dc Chara- drii formes (Aves) no litoral de Pernambuco, Brasil. Revista Brasileira dc Zoologia 18:657-672. Azi-:vi-;do-.Lmor, S. M.. M. M. Dias. M. b. Lakrazabal, W. R. Tfj.ino-JOnior and R. m. Lyra -Neves. 200 lb. Recapturas e recupcraydes dc aves migratorias no litoral de Pernambuco, Brasil. Arurajuha 9:33—12. AZEVEDO-JPNIOR. S, M.. M. E. I.ARRAZABA1. ANDO. PENA. 2004. Aces uquiilicas de ambientes antropieos (salinas) do Rio Grande do None. Brasil. Pages 255-266 in Aves mat in has e insulares brasileiras: biologia e conservayao (J. O Branco. Editor) Editpra Univali, ltajai. Brazil. Baker. A. J.. P. M. Gonz.ai.es. T. Piersma, L. J. Niles, I. I.. S. Nascimento. P. W, Aikinson. F. A. Clark, C. D. T. Minton, M. K. Peck, and G. Aarts. 2004. Rapid population decline in Red Knots: fitness consequences of decreased refueling rates and late arrival in Delaware Bay. Proceedings ot the Royal Society of London, Scries R 271:875-882, Barbu. Rt. E 2007. Seasonal abundance of shorebirds at Aracaju. Sergipe. Brazil Wader Study Group Bulletin 113:40-46. BARBU Rt. E. AND G. T HveneoaaRD. 2008, Seasonal occurrence and abundance ol shorebirds at Atulaia Nova Beach in Sergipe Stale, Brasil. Waterbirds 31:636-644. BARBIKRI, E. AND J, T. MF.NDONCA. 2005. Distribution and abundance of Charadriidac at Ilha Comprida Sao Paulo Brazilian. Journal ol Coastal Research 21:1-10. Barrier), E. and E. T Paes. 2008. The birds at Ilha Comprida beach (Sfto Paulo State, Brazil): a multivar¬ iate approach. Biota Neotropical 8:13-23, Barbilri. E.. J. T. Mendonca, and S, C. Xavier. 2003. Importance of Ilha Contprida (Sao Paulo State. Brazil) for the Sanderlings (Calidri.s alba ) migration. Journal of Coastal Research (Special Issue) 35:65-68. BE! TON, W. 2000. Aves do Rio Grande do sub Distribuiyao c biologia. Universiudc do Vale do Rio dos Sinos, Sao Leopoldo, Brazil. Bibby, C. J.. N. D. Blkciess. and D. A. Hill. 2000. Bird census techniques. Second Edition. Academic Press, London, England. BlRDLlEE International. 2011. Species tactsheet: Tryn- gites subruficollis. BirdLifc international, Cambridge, United Kingdom, www.birdlife.org Blanco, D. E.. P. Yorio. P. E. Pethacci. and G. Pdgnali. 2006. Distnbution and abundance of non-breeding shorebirds along the coasts of Buenos Aires Province, Argentina. Waterhirds 29:381-390. Chesser. R. T. 1994. Migration in South American, an overview of the austral system. Birdlife Conservation International 4:91-107. COSTA, E. S. and M. Sander. 2008. Variayao sa/onal de aves cosleiras (Charadrii formes e Ciconii formes) no litoral none do Rio Grande do Sul. Brasil. Biodiversi- dadc Pantpeana 6 ( i ):3-8. Crosse tv. R.. K. Karlson. and M. O'Brien. 2006. The shorebird guide. Houghton Mifflin Company. New York. USA. 50 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124, No. I, March 2012 Dajoz, R. 1983. Ecologia Geral. Vozes, Rio de Janeiro, Brazil. Fedrizzi. C. E.. S. M. AZEVEDO-JLINIOR, and M. E. Larrazabal. 2004. Body mass and acquisition of breeding plumage of wintering Calidris pu.iilla (Linnaeus) (Aves, Seolopacidae) in the coast of Pernambuco, north-eastern Brazil. Revista Brasileiru de Zoologia 21:249-256. Gianuca, N. M. 1983. A preliminary account of the ecology of sandy beaches in southern Brazil. Pages 413—419 in Sandy beaches as ecosystems (A. McLachlan and T. Erasmus. Editors). The Hague, Netherlands. Harrington, R. A.. P. T. Z. Antas, and F. Silva. 1986. Northward short-bird migration on the Atlantic Coast of southern Brazil. Vida Silvestre Neotropical 1:45-54. Lanctot, R. B„ D, E, Blanco, R. A. Dias. J. P. Isacch, v. a. Gill, J, B. Almeida, K Delhey. P. F. Petracci, G. A. Benckf., and R. A. Balbdkno. 2002. Conserva¬ tion status of the Buff-breasted Sandpiper: historic and contemporary distribution and abundance in South America. Wilson Bulletin 1 14:44 -72. Lara-Reshnde, S. l, and D. l. Leeuwenberg. 1987. Ecological studies of Lugon do Peixe. International Report 4. World Wildlife Foundation-WWP/US, District of Columbia. Washington, USA. Mata, J. R„ F. Erkk, and M. Rum BOLL. 2006. Aves de Sudamdrica: Guia de cumpo Collins. First Edition. Letemendia. Buenos Aires, Argentina. Morrison. R I. G. 1984, Migration systems of some New World shore birds. Pages 125-201 in Shorebirds— migration and foraging behavior (J. Burger and B. L. Olla. Editors). Plenum Publishing Corporation, New York, USA and London, United Kingdom. Morrison, R. I, G. and R. K. Ross. 1989. Atlas of Nearctic shorebirds on the coast of South America. Volume 1. Canadian Wildlife Service, Ottawa. On¬ tario, Canada. Morrison, R. I. G., R. K. Ross, and L. J. Niles. 2004. Declines in wintering populations of Red Knob in southern South America. Condor 106:60-7(1. Myers, J. P. and L. P. Myers. 1979. Shorebirds of coast.ii Buenos Aires Province, Argentina Ibis 121:186 20 Myers. J. P . M. A. Sallvberry, E. or rtz, G. Castro, l M. Gordon, J. L. Maron, C. T. Schick, E Tahiio.P AntaS. andT. Below. 1990. Migration mutes of Nes World Sanderlings ( Calidris alba). Auk I07.I72-1X( Parker HI. T A,. D. F. Stotz. and J. W. Fitzpatrick 1996. Ecological and distributional data bases Page- 1 13—436 in Neotropical birds: ecology and conserva- tion (D. F. Stotz, J. W. Fitzpatrick, T. A. Parker III, and D. K. Moskovits. Editors). University of Chicago Press. Chicago, Illinois. USA. Pit.RSMA, T. 2007. Using the power of comparison to explain habitat use and migration strategies of shorehird- worldwide. Journal of Ornithology 148:45-59. PlERSMA, T. AND A. LlNDSTROM. 2004. Migrating '.hnre birds as integrative sentinels of global environmental change. Ibis 146 (Supplement l):6l-69. Rodrigues, A. A. E. 2000. Seasonal abundance of Nearctic shorebirds in the Gulf of Maranhao. Brazil. Journal of Field Ornithology 71:665-675. SERRANO, I, L, 2(X)8. ( 'hallenges and advances at the Brazilian WUSRN sites. Omitologia Neotropical 19:329-337. Sick, H. 1979 Migrates de aves no Brasil. Brasil Flotestal 9:7-10. Teuno Junior, W. R„ S. M. Azevedo-Junior. and R VI Lyra-Mbndes. 2003. Ccnsos de aves migratorias (Charadriidae, Seolopacidae e Laridae) na Coroa do Aviao, Igarassu, Pernambuco, Brasil. Revista Brasi- leira de Zoologia 20:451-456. Vooren. C. M. and A. Chiaradia. 1990. Seasonal abundance and behavior of coastal birds on Cassino Beach, Brazil. Omitologia Neotropical 1:9-24. The Wilson Journal of Ornithology 1 24( 1 ):5 1 -56, 20 1 2 PLASTICITY OF HABITAT SELECTION BY RED-BACKED SHRIKES {LANIUS COLLURIO) BREEDING IN DIFFERENT LANDSCAPES FEDERICO MORELLI1 ABSTRACT.— Environmental parameters in different breeding habitats of Red-backed Shrikes ( Lanius collurio) in central Italy were examined at altitudes ranging from 0 to 1.200 m. I he most suitable habitats lor breeding were. ( I ) cultivated areas with hedgerows, and (2) high altitude grasslands. Similar population densities were recorded in both habitats (0.27 pairs/10 ha in farmland vs. 0.30 pairs/10 ha in meadows) and as were the number of fledged young per breeding pair (3.38 in farmland vs. 3.75 in meadows). The structural characteristics of ‘open space' and 'edge density' differed in the two breeding habitats. Use of species of trees and bushes for nesting depended on habitat type, but nests were in die more abundant thorny shrubs (blackthorn | Primus spinosa) in farmland and juniper \Junipems communis | in meadows). Red-backed Shrikes in farmland appear to prefer to nest in the most heterogeneous territories, those with the presence of uncultivated areas and shrub patches. Plasticity of habitat selection by the species was evident. Reiei\ed _/ June 2011. Accepted 14 September 201 1 Characteristics of breeding territories of Red- backed Shrikes ( Lanius collurio ) have previously been described in a number of studies (Cramp and Perrins 1993, Olsson 1995b. Lefranc and Worfolk 1997, Harris and Franklin 2000. Guerriere and Castaldi 2006, Casale and Brambilla 2009). The greatest density of breeding pairs in Europe is in farmland ecosystems (Cramp and Perrins 1993, Parkas et al. 1997, Golawski and Golawska 2007) with these landscapes providing highly varied ecological conditions for many bird species and 'beir prey (Golawski 2006). The Red-backed Shrike has been shown to prefer agricultural landscapes lhal are 'not intensively farmed' and are characterized by reduced functional heteroge¬ neity. This is especially true when there arc hedgerows and shrub patches that function as both key hahitat for many plants and animals, and wildlife corridors that enable dispersal and movement between habitats (Vermeulen 1994, Vcrmeulen and Opdam 1995). Habitats used for breeding have an essential n»le in the life-cycle of a bird, and many factors van influence the selection of breeding territories; two most important of which are minimization of predation (Cody 1985. Martin 1995. Roos -••021 and presence of adequate food (Martin '^7, Cramp and Perrins 1993, Golawski and Meissner 2008). The Red-backed Shrike is considered a I arm- land breeding species in the Marche Region ol Cen*ra| Italy (Pandolfi and Giacchini 1995, Forconi 2007. Morel I i et al. 2007). Hedgerows Jrc a familiar feature of most farmlands, and were OiSTcVA, University of Urbinp, Scientific Campus, 6j929 Urbino, Italy; e-mail: federico.morelli@uniurb.it traditionally planted as boundaries by both farmers and townspeople, albeit for different purposes. Currently, in areas which once had much greater coverage, they arc now only present as residual boundary lines of trees and shrubs, and are slightly >20 m in length and <5 m in width. Recent censuses indicate that Red-backed Shrikes may, however, use other habitat types for breed¬ ing in the Marche Region of central Italy (Forconi 2007, Morelli et al. 2007). Detailed information about habitats and eco¬ logical requirements of a species is needed if it is to be successfully conserved. This is especially true for the Red-backcd Shrike which, except for a few areas of relative stability (PEC'BMS 2009), has experienced a marked population decline in western and northern Europe over the last three decades. The Italian population has been esti¬ mated at 50,000 lo 120,000 reproductive pairs (Birdlife International 2004), but with a negative trend (Meschini and Frugis 1993, Pandolfi and Giacchini 1995, Dinetti 1997). The causes of this decline are poorly under¬ stood. and it has been suggested that reduction in suitable habitats, habitat modifications, use of intensive agriculture systems, decline in food resources, and climatic changes are the main reasons (Tucker et al. 1994. Yosef 1994, Fuller et al. 1995). Rapid changes in agricultural ecosystems can result in the loss of a bird species in just a few years. Thus, it is important lo examine the ability of a species to colonize and breed in areas other than agricultural farmland. This is because the capacity to breed or forage in a greater number of habitat types can improve the survival potential of birds which settle in a territory that includes an optimal proportion of 51 52 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 habitats suitable for successful breeding and survival (Soderstrom 2001, Golawski and Meiss¬ ner 2008). My objectives were to examine one population ot Red-backed Shrikes breeding in high altitude meadows (>900-1,100 m asl) and another in low- hill farmlands to compare environmental charac¬ teristics of these two habitat types. Methods Study Area. — The study was conducted during 2009 in the Marche Region of central Italy, which is characterized by a temperate climate, high spring and summer temperatures, and a marked summer drought (Tomaselli et al. 1972). The study area contained two breeding zones of 5,573 ha total (farmland: 43° 47' 27.81" N, 12" 39 39.91 E; 4,975 ha; high altitude meadows: 43" 37' 51.84" N. 12 45’ 48.18" E; 598 ha). Monitoring Frequency.— The study area was monitored for breeding Red-backed Shrikes every 5 days from early May to mid-July 2009. Occupied territories were surveyed at least once every 3 days during the breeding period to find nest sites (Bibby et al. 1997). Red-backed Shrikes are particularly sensitive to disturbance during nest building, egg-laying, and early incubation (Olsson 1995a. Tryjanowski and Kuzniak 1999), and the number of visits to each nesting site was kept to a minimum. Location of each nest site was recorded using a Global Positioning System (GPS) and numbers of fledged young per pair near a nest were recorded. The density of Red-backed Shrike territories in the two habitat types was calculated as breeding pairs per 10 ha. Environmental Parameters.- The typical size ot a Red-backed Shrike territory varies from I-? to 5-6 ha (Cramp and Perrins 1993. Lefranc I993~ Olsson 1995b. Harris and Franklin 2000). A 3-ha territory size around the nest site was standardized to enable territory preferences to be studied ArcGIS 9 software (ESR1 2009) was used to delineate a polygon with a 100-m radius around (AAVV w'„fnd a llmd'USC "'ap ":l0-00<» I i ." .' ) was llsccl to characterize the habitat in the polygons. The environmental parameters studied for each nest site were: (1) nest shrub: the plant species where breeding pairs placed their nest and w e very high (>5() items); (3) the altitude of the nesting site (m above sea level); and (4) com¬ position and percentage of land-use types around the nesting area (cultivated, uncultivated vine yard, shrub, forest, reforestation, grassland, river, buildings, and roads) The features and structural characteristics studied for each nest site were: ( I ) ‘open space' around the nesting site: percentage of cover of roads, grasslands, uncultivated land, and shrubs: (2) 'edge density' as the sum of the perimeters of all polygons in the buffer zone per number of land-use types/ 1 00 (as a surrogate of the habitat fragmentation level in the buffer zone); (3) road type (paved, unpaved); (4) distance from the nearest road (m); and (5) distance to the nearest building (m). Statistical Analysis. — Differences in the plani species used for nesting in the two habitat type-' were compared with Chi-square tests using standardized residual analyses to highlight the associations of plant species with habitat tjpes. A Mann- Whitney U- test was also performed to ex¬ amine differences in environmental parameters of breeding sites in farmland areas versus those in high altitude meadows. The structural cbaiacler- islics (open space and edge density) between the two habitats were further compared with a Chi- square test. I also performed a logistic regression analysis to identity which variables best explained the pre¬ sence or absence of Red-backed Shrikes in the farmland study area. Presence and absence of active nests were used as response variables, while composition and percentage of land-use types around the nesting area, structural characteristics, and road and building distance were the explana¬ tory variables. Internal validation was performed by calculating the non- parametric confidence interval of the AUC (Area Under the Curve1 values. The AUC generally ranges from 0,5 lor models with no discrimination ability to 1.0 l°r models with perfect discrimination. Data are pre¬ sented as means ± SE and all tests were conducted with SPSS lor Windows software. Version I ■ 11 (® SPSS Inc.. Chicago. 1L. USA. www.spss.com RESULTS I monitored and characterized 150 Red-backed Shrike nest sites in two breeding zones: 132 m farmland (mean altitude = 347.2 ± 138.8 m asl) and 1 8 in high altitude meadows (mean altitude = 837.6 — 1 15.5 m asl) (Fig. 1). The density o! breeding Morelli • HABITAT SELECTION BY RED-BACKED SHRIKES 53 FIG. 1. Nest site localization in two different Red-backed Shrike breeding habitat types. pairs per 10 ha was 0.27 for farmland and 0.30 for high altitude meadows, respectively. The number of fledged young per pair was similar in the two habitats: 3.38 ± 1 .05 (n = 31) for farmland and 3.75 - A-86 (// = 12) for meadows. These differences were not significant (U = 148. P = 0.282). Plant Species Used as Nest Sites.— AW nest sites were in shrubs. Plant species used differed in the two breeding habitats (xJ - 71.3, P < 0.05) '■'•ith blackthorn ( Prumis spinosa) being most t48.5% ) used in farmland followed by dog rose canirue, 25.8%), elm-leaf blackberry (Paints *lnifolius\ 12.1%). and common hawthorn t Crataegus monogyna ; 8.3%) (Table 1). Plant ■species used most frequently in high altitude meadows were juniper (Juniperus communis ; 50%), blackthorn (22.2%), elm-leaf blackberry (11.1%). European beech ( Fagus sylvaticcr, I 1.1%), and dog rose (5.6%) (Table 1). Land-use Cover in Farmland and High Altitude Meadow Breeding Sites.— The percent cover of the types of land-use around nest sites differed between the two breeding habitats studied. Up to six habitat types (mean = 3.7) were recorded for farmland with cultivated land, shrubs, and grass¬ land dominating (Table I). Five habitat types (mean = 2.9) were recorded in meadows, where grassland and forest prevailed (Table 1). Characteristics and Structural Differences Be¬ tween Breeding Habitats.— Open space and edge TABLE 1. Plant species used for nest sites by Red-backed Shrikes in farmlands and high altitude meadows in central Italy. _ Farmland High altitude meadows Species Name n % Abundance n % Abundance Clematis vitalha Clematis 3 2.3 Low Cnaagus monagxna Common hawthorn 11 8.3 Medium Pvgus sylvatica Beech 2 11.1 Low hmipena communis Juniper 9 50.0 Very high 1 11 Hums spina-christi Christ’s thorn 1 0.8 Very low Paimis spinosa Blackthorn 64 48.5 Very high 4 22.2 Medium fobinia pseudoacacia Black locust 1 0.8 High R°sa canina Doe rose 34 25.8 High I 5.6 Medium Rubus ulmifolius Elm-leaf hackberry 16 12.1 High 2 11.1 Medium Sambucus nigra Elderberry 2 1.5 Medium 54 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. I. March 2012 TABLE 2. Characteristics ot environmental parameters of farmland and high altitude meadows used by Red-backed Shrikes for breeding in central Italy. Characteristics Farmland (n = 132) Meadows (n = 18) u p Building, % Cultivated. % Vineyards, % Forest. % Grasslands, % Uncultivated and shrubs, % Water. % Roads, % Landuse categories, number Near road, m Near building, m 3.4 ± 6.8 57.4 ± 26.1 0.9 ± 3.6 6.7 ± 13.2 14.2 ±21.5 1 1.0 ± 19.9 0.3 ± 1.4 6.0 ± 3.2 3.7 ± 1.2 30.6 ± 43.7 186.4 ± 136.7 0.8 ± 2.6 2.4 ± 3.9 0.0 ± 0.0 27.6 ± 28.4 65.2 ± 27.9 0.6 ± 2.1 2.2 ± 8.4 1.2 ± 2.6 2.9 ± LI 145.6 ± 123.6 781.6 ± 475.2 862.0 64.0 1.080.0 490.0 189.0 818.5 1.052.5 362.0 692.5 369.5 355.5 0.028 0.000* 0.184 0.000* 0.000* 0.016 0.070 0.000* 0.003* 0.000* 0.000* * Significant at P < 0.01. density values differed between the two areas used for breeding (Z2 = 47.3, P < 0.05). The percentage of open space in high altitude meadows (67.0 ± 28.2%) was higher than in farmland (31.3 ± 26.5%). In particular, the main types of land-use cover functioning as open space in farmland were roads, and uncultivated land, while in meadows the main type was grassland. The edge density value in farmland (93.4 ± 45.8) was higher than in meadows (50.9 ± 54.3). There were significant structural differences between the two habitats in terms of distance of nests from the nearest road and building (the shortest distance was in farmland) (Table 2). The model produced using stepwise logistic regression analysis was based on 35 nest sites selected at random from the farmland breeding habitat and 35 other sues where the species had not been present Oeal absence) within the same studied farmland Only five variables had a substantial effect on probability of the occurrence of the species in farmland (Table 3). These were: edge density open space, uncultivated land, and shrubs forests’ and grassland. The first three had a positive effect on the probability of the occurrence of Red- backed Shrikes, whereas the other two had a marginally negative effect. DISCUSSION Red-backed Shrikes seem to be much more cc mon ,n low-altitude farmland than in mount environments in the Marche Region of central It (FoKon, ,007. Morel li and Pandojfi 2009. Mon -01 la), as is also the case in other parts oft country. However, the results of my study indie- •bat Red-backed Shrikes can us/very 7fte environments and select, if available, high meadow areas for nesting. 1 he two populations studied were in very dif¬ ferent altitude zones with farmland birds breeding at a mean elevation of 350 m. while the mea¬ dow- population was at an elevation >800 mask However, population densities of shrikes were similar in both areas, as were numbers of fledged young per pair. 1 hese results highlight the ecological plasticity of Red-backed Shrikes in selecting breeding habitats, as this species can use different types ol landscape in central Italy for breeding. This suggests these two habitat types meet the eco¬ logical requirements of Red-backed Shrikes. Farmlands where Red-backed Shrikes were detected were characterized by higher landscape heterogeneity (edge-density values, roads and buildings in the vicinity), reducing risks from predators or providing natural hunting perches (Lefranc 1993, Yosef and Grubb 1994, Lefranc and Worfolk 1997, Bechet et al. 1998, Tryjanow- ski et al. 2000. Roos 2002. Roos and Part 2004 1 Variables that appear to have made a major con¬ tribution to the suitability of farmland as breeding TABLE 3. Logistic regression of presence and absence ol breeding Red-backed Shrikes in farmlands (only significantly associated variables are shown). Variable B SE P Edge density 0.05 0.012 <0.001 Open space 45.12 17.67 0.01 1 Uncultivated and shrub 1.13 0.525 <0.001 Forest -1.12 0.436 0.012 Grassland -21.42 9.21 0.05 Morelli • HABITAT SELECTION BY RED-BACKED SHRIKES 55 habitat for Red-backed Shrikes were edge density, which indicates a certain functional heterogeneity i Benton et al. 2003), open spaces, uncultivated land, and shrubs for nesting. In contrast, the spe¬ cies does not seem to favor farmland with forest patches or grassland cover. Meadows frequented by Red-backed Shrikes were less fragmented than the farmland. They also contained virtually no roads or building structures, but abundant shrubs were uniformly distributed across open grasslands. These high altitude meadows can provide another important resource to maximize foraging ecology of shrikes in the increased availability of 'open space', as open grasslands at high altitude are optimal for finding and capturing prey (Fernandez-Juricic et al. 2004, Golawski 2006. Morelli 2011b). Recent studies reveal that meadows, in comparison with other habitat types, support the highest number and biomass of invertebrate prey of Red-backed Shrikes (Golawski and Meissner 2008). ‘Open space’ had different features in each type ol breeding habitat (mainly roads in farmland and grassland in meadows). The Red-backed Shrike used different kinds of shrubs in which to nest and perch in the two habitats; generally, the most abundant, suitable species (blackthorn in farmland and juniper in meadows). These characteristics, which highlight the Red- backed Shrike's relative ecological plasticity, C0l,ld be important when defining 'suitable habitat’ and ‘habitat availability' for the species, ihese findings should be used to identify new potential breeding sites and expansion areas for Red-backed Shrikes in central Italy. Future studies focused on breeding success of shrikes in different habitat types may be necessary if wc are to obtain reformation required to understand the biological requirements of this species. This information lna> have an important role in the success ol bird conservation programs (Hoffmann and Greet Salem 2003, Morelli et al. 2007). ACKNOWLEDGMENTS Wc 'bank Maurizio Saltarelli and Chiara Tagnani for *lelp in the field, and Marco Girardello and Maria Balsamo 1('r valuable suggestions on ihc text. A special note ot ’banks is owed to the anonymous reviewers lor suggesting '■Veral improvements in the original manuscript. LITERATURE CITED A/VVV. 2001. Land use map 1:10.000 Marche Region. Regional Cartographic Office, Marche. Ancona, Italy. Bechet, A.. P. Isenmann, and R. Gaudin. 1998. Nest predation, temporal and spatial breeding strategy in the Woodchat Shrike Lanins senator in Mediterranean France. Acta Oeeologia 19:81-87. Benton, T. G„ J. A. Vickery, and J. D. Wilson. 2003. Farmland biodiversity: is habitat heterogeneity the key? Trends in Ecology and Evolution 18:182-188. Bibby, C. J.. N. L). Burgess, and D. A. Hill. 1997. Bird census techniques. Academic Press. London. United Kingdom. BihdLife International. 2004. Birds in Europe: popula¬ tion estimates, trends and conservation status. BirdLife Conservation Series Number 12. BirdLife Internation¬ al. Cambridge, United Kingdom. Uasai.E, F AND M. BRAMBILLA. 2009 Averla piccola. Lcologia e conserva/ione. Fondaztone Lombardia per PAmbiente c Regione Lombardia, Milano. Italy. Cody. M. L. 1985. Habitat selection in birds. Academic Press, New York, USA. Cramp, s. and C. M. Perrins (Editors). 1993. The birds of the Western Palearctic. Volume 7. Oxford University Press. Oxford, United Kingdom. Dine rn. M. 1997. Averla piccola. Lanius colturio. Page 308 in Atlante degli uccelli nidificanti in Toscana ( 1982-1992) (G, Tell ini Florenzano. E. Areamone. N. Baccetti, E. Meschini, and P. Sposimo. Editors). Monografie I . Quaderni del Museo di Storia Naturale di Livorno. Italy. ESR1. 2009. ArcGIS Desktop. Release 9. Environmental Systems Research Institute. Redlands. California, USA. Parkas, R„ R. Horvath, and L. Pasztor. 1997. Nesting success of Red-backed Shrike, Lanius collurio in a cultivated area. Ornis Hungaricu 7:27-37. Fernandez-Juricic. E.. J, T. Erk iisen, and A. Kacelnik. 2004. Visual perception and social foraging in birds. Trends in Ecology and Evolution 19:25-31. Forconi, p. 2007. Averla piccola, Pages 270-271 in Atlante degli uccelli nidificanti nella provincia di Ancona (P. C.iacchini, Editor), Provincia di Ancona, LX Set lore Tutela dell' Ambicnie— Area Flora c Fauna, Ancona. Italy. Fuller, R- J-. R. D. Gregory, D, W. Gibbons, J. H. MaRciiant, J. D. Wilson. S. R. Bailue. and N. Carter. 1995. Population declines and range contrac¬ tions among lowland larmland birds ill Britain. Conservation Biology 9:1425-1441. Golawski. A. 2006. Diet of the Red-backcd Shrike Lanius collurio in the agricultural landscape of eastern Poland. Notatki Omilologic/.ne 47:208-213, Golawski. A. and S. Golawska. 2007 Habitat preference in territories of the Red-backed Shrike Lanius collurio and their food richness in an extensive agriculture landscape. Acta Zoologica Academiae Seientiarunt Hungaricae 54:89-97. GOLAWSKJ. A. AND W. Meissner. 2008. The influence of territory characteristics and food supply on the breeding performance of the Red-hacked Shrike (Lanius collurio 1 in an extensively farmed region of eastern Poland. Ecology Research 23:347-353 Guewuere. G. and A. Cast aldi. 2006, Curarterisf/che del sito di nidificazione, densita e biologi a riproduttiva 56 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. 1. March 2012 dell Averla piccola. Lanius collurio. in Italia centrale. Avocetta 29:5-1 1. Harris. T. and K. Franklin. 2CKK). Shrikes and bush- shnkev Christopher Heltn. London. United Kingdom. Hoffmann, J. and J. M. Greef. 2003. Mosaic indicators— theoretical approach for the development of indicators lor species diversity in agricultural landscapes. Agriculture, Ecosystems and Environment 98:387 Lefranc, N. 1993. Les Pies-grieches d'Europa, d'Afrique du nord et du moyen-Orient. Delachaux ct Niestle S.A.. Lausanne, Paris, France. Lefranc. N. and T. Worfolk. 1997. Shrikes. A guide to the shrikes of the world. Pica Press. Sussex. United Kingdom MAT’J- E ,y87- Food a-s a limit on breeding birds- a hte-h.story perspective. Annual Review of Ecology and Systcinatics 1 8:453-487. Martin. T E. 1995. Avian life history evolution in relation to nest sites, nest predation, and food. Ecological Monographs f»5: 101 — 127. MeschinI; E. and S. Hr i jus, |9Q3. Atlante degli Uccelli null I icant i in Italia. (Breeding birds in Italy). Supple- memo Ricerca Biologia della Selvaggina 20:218. orelli F. 2011a. Distribuzione altitudinale dei siti nproduttiv, di Averla Piccola Lanins collurio' in ambient i agricol, della provineia di Pesaro-Urbino Itaha centrale. (Altitudinal distribution of breeding 70:91-95 aCkCd CCmral ,taJ>* P'c™ Morelu. F 201 lb. Importance of road proximity for the ne^ site selection of the Red-backed Shrike Lanius . ’ : ™ agricultural environment in central Italy Journal of Mediterranean Ecology | |-->|-->y rANDOm- 2009 H“blmi * oMifio. zione de | Averla piccola lamas collurio nellc “ .CCn,ralc (Breeding habitat of Red- backed Shrike in the Marche Region, central Italy) ™s;7?nova- «-< - <$& M°R2TO7 u M' PAN“"'H- S PESARKI- «* E- B.ond, 2007 Using monitonng data and habitat variables to 'h.C d'STnbution of bird species in the Marche Region (Italy). Rtosociologia 44 (Supplement I ): 1 27- OL'SON. V 1995a. The Red-backed Shrike Lanius Collurio m southeastern Sweden: breeding biology. Ornis Svecica b:IOI-l 10 OLSSON. V. 1995b. The Red-tacked Shrike Lanius caUario SvedcI Sn,5”'11'" hahi'al a"d Pandolfi, M. and P. Giacchini. 1995. Avifauna neili provineia di Pesaro. Centro Stampa Amminiflnzi- Provincialc di Pesaro e Urbino, Assessorato Ambbi Pesaro e Urbino. Italy. Pan-European Common B/rd Monitoring Schrii (PECBMS). 2009. The state of Europe's comn birds 2008. CS(J/RSPB. Prague. Czech Repubil. http://www.birdlife.cz/index. php?ID= 1 320 Roos- S. 2002. Functional response, seasonal decltoc •>. landscape differences in nest predation risk Oecukio 133:608 -615. Rons. S. AND I PART. 2004. Nest predators affect sprail dynamics of breeding Red-backed Shrikes hmw collurio. Journal of Animal Ecology 73:117-127. Salem, B. B 2003. Application of CIS to biodivers't) monitoring. Journal of Arid Environments 54:91- 114. Soderstrom, B. 2001, Seasonal change in Red-backed Shrike Lanius collurio territory qualitv-thc role of net predation. Ibis 143:561-571. TOMASELLI, R • A. BalDUZ/1. AND S. Fiupf.u.o. 1972. Carta Bioclimatica d’ltalia ScaJa L2.(MMJ,000. istililto di Botanica-Universitii di Pavia. Ministero Agricoltun c Foreste: Collana Verde 33. Italy. I RYJANowski, P. AMDS. Kuzniak. 1999. Effect of research activity on the success of Red-backed Shrike, 'lawn' collurio nests. Ornis Fenniea 76:41-43. Trvjanowski, P,. s. Kuzniak. and B. Diehl. 2(H«. Dee- breeding performance of Red-backed Shrike, larim collurio. depend on nest site selection? Ornis Fetmi.e 77:137-141. Tuc ker. G„ M. Heath, L. Tomialojc'. and R. F. A. Grimmett. 1994, Birds in Europe: their conservation slatus. Cambridge University Press. Cambridge. Unit¬ ed Kingdom. VERMEULEN. H. J. W. 1994. Corridor function of a road verge for dispersal of stenotopic heathiand ground beetles Carabidac. Biological Consenation 69:339- 349. VERMEULEN. H. J. W. AND P, Opdam. 1995. Effectiveness oi roadside verges as dispersal corridors for small ground-dwelling animals: a simulation study. Lind- soape and Urban Planning 31:233-248. 't osei , R. |Q94 Conservation commentaiy: evaluation ot the global decline in the true shrike, family Lanidac. Auk 111:228-233. Yosef. R, .and T. C. Grubb. 1994. Resource dependent.' and territory size in Loggerhead Shrikes Limit's luiJovicianus. Auk 111:465-469. The Wilson Journal of Ornithology 124( 1 ):57-65, 2012 TERRITORY DISTRIBUTION AND HABITAT SELECTION OF THE SERRA FINCH ( EMBERNAGRA LONGICAUDA ) IN SERRA DO CIP6, BRAZIL GUILHERME H. S. FREITAS' 2 AND MARCOS RODRIGUES' ABSTRACT.— The near-threalened Serra Finch ( Embemagra longicauda) is restricted to the main mountain ranges in eastern Brazil inhabiting compos rupestres (rocky fields). We mapped 17 mated pairs in a 138-ha area within Serra do Cipo National Park; a density of 0.25 adults/ha. Estimated average territory size varied from 2.52 ± 0.77 ha (95% kernel ) to 3.35 • 0.90 ha f 100% minimum convex polygon). The distance between territory centers of neighboring pairs was 162.38 ± 28.93 m, The overlap between neighboring territories was 15.3 ± 5.9% (95% kernel) and 2.0 — 2.3 r (polygon metho ). Pairs remained together throughout the year in the same territories and defended these against intruding neighbors. Analyses of habitat selection indicated preference for woodland and scrubland habitats associated with humid valleys, whi e grasslands were avoided The Serra Finch used the available habitats more than expected from random at ditferent spatial sales. Our data identified habitats that should be priority for conservation of the Serra Finch. Received 8 Septan er _ •T cepted 24 September 201 1. The Serra Finch ( Embemagra longicauda) is a poorly understood species, known for more than a century based on only two specimens from an unspecified locality in South America (O’Brien l%S. Mattos and Sick 1985). Recent studies have found it to be restricted to mountaintops in eastern Brazil, mainly in the states of Minas Gerais and Bahia, and particularly Hie Espinha^o Range, Sena da Mantiqueira and Serra do Caparad iVasconcelos 2008). This species may be easily observed on mountains above 900 m. frequenting habitats on quartzite, gneiss, and iron rich soils. Thus species typically places its nest on rocky outcrops (Hoffmann et al. 2009, Rodrigues et al. tends to forage in pairs for arthropods and B«hy fruits among herbs, grasses and bushes 'Hoffmann et al. 2009). and pairs arc known to Ning a duet < Freitas and Rodrigues 2007), a typical behavior of territory defense (Catchpole and Slater 1995). The Serra Finch has been categorized as near- Oireatened due to a perceived population decline possibly resulting from significant habitat loss, primarily due to cattle ranching, land conversion, and intensive mining activity (BirdLife Interna- honal 2011). Thus, there is a need of focused ecological studies of the species to identify its Habitat requirements (Stotz et al. 1996. BirdLite International 2011). Accurate assessment o! ' Gboratorio de Ornilologia, Departamenlo de Zoologia. H'tiiuto dc Ciencias Biological, Univcrsidade Federal de Minas Gerais, CP 486. 31270-901. Bclo Horizonte. Minas Gerais. Brazil. Corresponding author; e'ma‘l: guilhermehsfreitas@gmail.com habitat requirements, population size, and den¬ sity are paramount to undertaking any successful conservation measures (Bibby et al. 2000). The objectives of our study were to: (1) quantify and map all territories of Serra Finches in a study site within Serra do Cipo National Park in south¬ eastern Brazil, and (2) identify specific habitats favored by this species. METHODS Study Aren. — Our study was conducted in a 138-ha area near Indaifi Stream, in the ’Alto do Palacio’ region (19 15' S, 43 31' W) in the northern part of Serra do C’ipo National Park in the southern portion of the greater Espinha^o Range of southeastern Brazil (Fig. I). Alto do Palacio is near the ridge of the eastern slope of the Serra do Cipo Mountains at an altitude of 1 .280 to 1,380 m. This region is humid throughout the year, even in the dry season, and is characterized by frequently misty weather conditions (Ribeiro et al. 2009); the bird community is closely associated with the Atlantic Forest (Rodrigues et al. 2011). The study area is a mosaic of habitats traversed by numerous small valleys. Seven habitats were identified. (I) Rocky outcrops, referring to areas wilh soils derived from quartzite, that are domi¬ nated by several species of herbs, shrubs (such as Bromeliaceae. Orchidaceae. Velloziaceae. and Cactaceae), and small trees up to 3 in in height, including Eremanthus erythropappus and E. cro- tomides ( Asteraceae). (2) Dr y grasslands, domi¬ nated by Lagenocarpus tenuifolius (Cype raceae), Panicum lore um (Poaceae). and Paepa Ion thus spp. 57 58 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 (Eriocaulaceae). (3) Wet grasslands, dominated I Lagenocar/ius rigidus in marshy areas close watercourses. (4) Dry scrubland, which has son plants that also occur in the rocky outcrop habit: but which is dominated by Coccoloba acrosi choides (Poligonaceae) and Aulonemia effic (Poaceae) in sandy soil with substantial gravel some hilltop areas. (5) Wet scrubland, dominate by small trees up to 2 m in height, such ; Tibouchina spp., Lavoisiera imbrkata (Melasti mataceae), and Baccharis ilatiaiae (Asteraceae) i marshy areas. (6) Riparian woodlands, charactei ■zed by dense vegetation and dominated by 4 t 10-m tall trees including Miconia chartace, (Melastomataceae) and Hololepis peduncular (Asteraceae). (7) Candeial , a specific formatioi of dense groups of 3 to 5-m tall Eremanthm spr (Asteraceae) trees, known as ‘candeia’. The tern dry is not associative of arid conditions bu differentiates from saturated soil conditions of the wet habitats. processing individuals (—80 hrs), and locating birds. We captured Serra Finches using mist nets and banded each with unique combinations of colored leg bands to allow individual identification and numbered metal leg bands supplied by Centro Nacional de Pesquisa para Conserva^ao de Aves Silvestres (CEMAVE/IBAMA. license # 1 161/3 and I 161/5). Surveys were conducted throughout the study area. Locations of Serra Finches were recorded using a hand-held Garmin (Olathe. KS USA) global positioning system (GPS) unit lo within 9 m accuracy. All GPS locations corre¬ spond to pairs because finches were always detected in pairs. Locations were recorded every H> m of a pair's movements (not caused by observer presence) until lost from view. Habitat type was recorded where birds were observed. Territory Distribution.— We took GPS loca¬ tions lor all marked and unmarked pairs detected in the study area. Pairs were detected mainly while singing a duet from a perch, which was interpreted as evidence of a particular mated pair's territory. Locations were recorded through¬ out the year (dry and wet seasons), as we did not Freitas and Rodrigues • TERRITORY SIZE AND HABITAT USE OF SERRA FINCH 59 observe significant seasonal changes in spatial distribution and general territory defense of mated pairs. We performed spatial analyses using Arc- View G1S Version 3.2 (ESR1 1998). The distance between nearest neighboring pairs was calculated from the arithmetic mean of the locations of each mapped pair. We calculated the density of in¬ dividuals in the study area by mapping territories ■ Bibby et al. 2000). Territory size was calculated using the Home Range Extension Version 1.1 (Rodgers and Can" 1998) We used the following estimators: 95% fixed kernel using the least-squares cross-valida¬ tion smoothing parameter (Worton 1989. Seaman and Powell 1996): 50% fixed kernel for the core areas (Powell 2000); and 100% minimum convex polygon (MCP; Mohr 1947). We chose the kernel method because it is the most recommended (Laver and Kelly 2008) and MCP for comparison because it is the most used and emphasized the boundaries of space used (Kernohan et al. 2001), This is important for territory character ization based on the concept of use of an exclusive area ( Pitelka 1959). We measured territory size only for pairs with >30 locations to avoid im¬ precision and bias in the size estimates (Seaman ctal. 1999, Kernohan et al. 2001). Habitat Selection.— Wo analyzed habitat selec¬ tion by comparing habitat use and that expected based on availability. We used the described ha¬ bitat classifications for availability and delineated habitats based on satellite images from Google forth (2010) in combination with direct observa¬ tions iri the field and measured with X-tools extension for ArcView Version 3.2 (ESRI 1998). We calculated habitat selection using three approaches. First, we compared the number ol all $erra Finch locations within each habitat with the Proportion of available habitats in the study area. Lbi-square tests were used to evaluate the null hypothesis that actual use of different habitat 'ypes is directly proportional to their availability 'Neu et al. 1974). Bonferroni confidence intervals 11 = 0.05) were calculated from the observed Proportions of habitat use to identity which ha- bitat types were selected (Ncu et al. 1974, Byers and Steinhorst 1984). Second, we compared the Proportions of habitat within the 95% kernel territory boundaries to proportions of available habitats in the study area (second-order habitat election sens,, Johnson 1980). Third, we com¬ pared the proportion ol locations in each territory 10 the proportion of available habitats within the 95% kernel territory boundaries (third-order habitat selection sensu Johnson 1980). We used compositional analysis (Aebischer et al. 1993) for the second and third-order approach. This analysis used the Wilks' lambda statistic test for overall differences in habitat use. Comparisons of particular habitat types w'ere made with paired /-tests (Aebischer et al. 1993) if the Wilks' test suggested differential habitat use. Significance was set at P < 0.05. Habitat types were ranked from most to least selected using a matrix of mean and standard deviation of log ratio differences for all habitat types it selection was significantly non random. Missing values (zero) in the data matrix were replaced by O.OOI. The minimum number ot individuals needed for compositional analysis is six (Ae¬ bischer et al. 1993). Thus, all Sena Finches with >10 locations were used and those with <10 locations were excluded from the analyses. Leban et al. (2001) noted that compositional analysis is primarily affected by the number of animals sampled and did not vary much with number ot locations (as few as 10 observations). We conducted the habitat selection analyses using Resource Selection for Windows, Version 1.0 (Leban 1999). RESULTS Territory Distribution.— Twenty-one individual birds were marked with colored leg bands and 17 were relocated distributed among 1 1 mated pairs. The four individuals not relocated included a juvenile and three possibly adult floaters. Six additional mated pairs of unmarked individuals were also detected (during at least 2 different days) occupying areas between some ot the marked pair’s territories. Those pairs were reliably identified because the adjacent pairs often sang duets at the same time. Thus, at those moments it was possible to hear up to three different pairs. Territories of 17 total pairs were mapped by 3 18 GPS locations in the entire 138-ha area during the study period (Fig. 1). This was a density of adults with established territories of 0.25 birds/ha or 0.12 pairs/ha. Distance between territory centers was 162.38 ± 28.93 (range = 119.95-233.85 m. n = 17). Territory size estimates (mean ± SD) of five pairs that were located most often ranged from 2.52 ± 0.77 ha (MCP) to 3.35 ± 0.90 ha (95% kernel) (Table I). The area of common use between territories (MCP) was 991.27 nr for pairs A (4.68% of 60 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 TABLE 1 . Data for nine best sampled pairs of the Serra Finch in Serra do Cipo National Park. Brazil. Values of 100% minimum convex polygon (MCP); and by 95% and 50% kernel in ha. Bird pair Number of locations Sampling days Sampling period" 100% MCP 95% Kernel 50% Kernel A 43 6 435 2.11 3.29 0.85 B 57 14 412 3.30 3.62 1.02 C 52 9 273 2.68 2.84 0.63 D 44 6 168 3.09 4.69 1.16 E 14 3 32 0.78 2.66 0.75 F 13 3 113 1.46 4.87 1.23 G 30 4 169 1.42 2.32 0.55 H 12 3 1 12 1.24 4.78 1.15 K 16 4 169 0.88 1.85 0.40 Number of days elapsed between the first and the last locations. territory) and B (3.00%), and 19.34 nr for pairs C (0.07%) and D (0.06%) (Fig. I). The estimated overlap in area between territories (95% fixed kernel) was 7,204.59 nr for pairs A (20.72%) and B ( 1 9.35%), and 3,673.5 1 nr for pairs C ( 1 3.09%) and D (7.94%) (Fig. 2). The eore areas over¬ lapped for 1,155.62 m2 for pairs A (12.03%) and B (10.99%), and did not overlap between pairs (’ and D. approaches. Dry and wet grassland were essen¬ tially avoided by all individuals (Table 3), and we excluded them from the compositional analysis. The proportion of habitat use based on 95% kernel territories in the second-order approach was non random relative to that available in the study area (A = 0.13, df = 4. X2 = 18.30, P < 0.05). A ranking matrix ordered the habitat types in the sequence: wet scrubland > riparian woodland > rocky outcrop > candeial > dr. scrubland (Table 4). Serra Finch territories had significantly more wet scrubland and nparian woodland than dry scrubland. Habitat use of territory locations versus 95% kernels in the third-order approach significantly differed from random (X = 0.20, df - 4. X: = 14.41, 11 < 0.05). The matrix ranked the habitat in the order: candeial > riparian woodland > rocky outcrop > dry scrubland > wet scrubland. Candeial was significantly more used than rocky outcrop, dry scrubland, and wet scrubland, the last was significantly less used than riparian woodland (Table 4). DISCUSSION We did not observe any changes in the composition of mated pairs or in their territories, indicating these birds have annual site fidelity and a socially monogamous mating system. All pairs were observed singing duets on perches, appar¬ ently defending their territories throughout the year, even outside the breeding season. We observed agonistic events („ = 4) on territory boundaries, when two pairs stayed in close proximity performing a unison duet, and ending in a physical confrontation on two occasions. One of those, between pairs A and B. occurred within the MCP overlap area. Habitat Selection.- The 3 1 8 locations recorded in the habitat types were not distributed as expected from availability in the study area (I1 < 0.001, x2 = 686.82, df = 6; Table 2). The candeial. rocky oulcrops, wet and dry scrubland, and riparian woodland habitats were used more t an expected by chance as the proportions of expected use were below the Bonferroni confi¬ dence intervals for observed use. Use of dry and we grassland habitats was lower than what could be expected by chance (Table ~>) We used the location data of nine Serra Finches which had >10 locations (Tables 1-3) to examine hab,lat Wfecllon by (he .second and third orZ territory Distribution. — The number ot — Serra finches that we found is high compared with other species that inhabit open areas in central Brazil. Our results revealed higher densi¬ ties than 1 1 species studied by Braz (2008). including the closely related Black-masked Finch ( Cotyphuspiza melanntis) (0.23 individuals/hai and Wedge-tailed Grass Finch [Emberizoides Iwrbicola) (0.15 individuals/ha). Silva (2008) reported 1 1 individuals/ha for Serra Finch. Hoff¬ man (201 1 ) estimated a density of 0.44 ± 0.08 individuals/ha for Gray-backed Taehuri ( Polystic ■ tus supereiliaris). a tyrant-flycatcher with a range distribution similar to the Serra Finch (Vasconce- los 2008). Population density is highly dependent on the quality of a particular ecosystem (Makar- ieva et al. 2005, Johnson 2007). Our density data are for a population of the Serra Finch occurnng in an undisturbed area, which may serve as a baseline lor comparison with future studies o! this species. Mattos and Sick (1985) reported 400 in as the distance between pairs of Serra Finch, which is higher than the average distance estimated for the birds at our study site. Our results were derived Irom small sample sizes, but it is the most accurate estimate available, particularly given Freitas and Rodrigues • TERRITORY SIZE AND HABITAT USE OF SERRA FINCH 61 TERRITORY D TERRITORY C £ 1 4 . . i 95% kernel □ 50% kernel g 100% MCP ^neighbor 95% kernel (overlap) ^ Candeial °ry grassland ■ Wet grassland 1 1 1 Rocky outcrop iziDry scrubland Wet scrubland m. Riparian woodland w 90 m s FIG- 2- Habitat components of the 95% kernel territories ^onal Park. Brazil. of four pairs (A-D) of the Serra Finch in Serra do Cipo Mottos and Sick (1985) did not specify how 'heir values were calculated. The mean MCP territory size of Serra Finch "as relatively small compared to published territory estimates for other species that inhabit ,he Cerrado Biome including the Rufous- fronted Thornbird ( Phacellodomus rufifrons ) (3.43 ± 0.62 ha; Rodrigues and Carrara 2004), Suiriri Flycatcher (Suiriri suiriri ) (14 ± 1.9 ha; Lopes and Marini 2006), Chapada Flycatcher ( S . isler- orum ) (1 1.2 ±0.6 ha; Lopes and Marini 2006). Crested Black Tyrant (Knipotegus loplwtes) (7.3 ± 0.57 ha; Ribeiroet a I. 2002), Cinnamon Tanager ( Schistochlamys ruficapillus) (8.4 ± 1.9 ha; 62 THE WILSON JOURNAL OF ORNITHOLOGY • Voi 124. No. 1. March 2012 TABLE 2. Habitat selection of Serra Finches in Serra do Cipo National Park, Brazil. The proportion expected was calculated from the relative proportion of each habitat available within the study area, and the proportion observed in relation to 318 locations obtained for 17 pairs. Habitat selection is indicated for differences <0.05 significance; positive {+) for expected proportions below the Bonferroni confidence interval, and negative (-) for expected proportions above the interval. Habilal Proportion expected Proportion observed P, Confidence interval Selection Candeial 0.003 0.053 0.039 < P, < 0.068 + Dry grassland 0.793 0.019 0.010 < P, < 0.028 Wet grassland 0.079 0.025 0.015 < P} < 0.352 _ Rocky outcrop 0.047 0.366 0.334 < P4 < 0.397 + Dry scrubland 0.016 0.072 0.055 < /J, < 0.089 -f Wet scrubland 0.038 0.234 0.207 < Pf, < 0.262 + Riparian woodland 0.022 0.231 0.204 < P1 < 0.259 + Domingues and Rodrigues 2007), Shrike-like Tan- ager (Neothraupis fascia ta) (4.3 ha; Alves 1990), Henna-capped Foliage-gleaner (Hylocryptus rectir- ostris) (2.9 ± 1.4 ha; Faria el al. 2007), C.ray- hacked Tachuri (4.3 ± 1 .2 ha; Hoffmann 2006), and Cipo Canastero (Astltenes luizae ) (4.0 ± 2.6 ha; Freitas 2011). These studies all used similar methods, without radiotelemetry (except Freitas 2011). The home range area of those species could be larger than reported (Anich et al. 2009). The Serra Finch appears to be similar to most tropical bird species, maintaining a long-term pair bond and a territory throughout (he year (Stuehbury and Morton 2001), at times making it difficult to discern the territory space of the total home range area (Lopes and Marini 2006). The territorial behavior was illustrated by the annual site fidelity associated with duet song performance, and mini¬ mum overlap of the used area visualized by the MCP. Habitat Selection. — Johnson ( 1980) proposed a natural order to the process of habitat selection according to the scale of observation. First-order selection identifies the physical or geographical distribution of a species which, for the Serra Finch, is the mountaintop habitat complex that occurs throughout eastern Brazil (Vasconcelos 2008). Second-order selection delineates the home- range area of an individual or social group. Serra Finches did not establish territories at random and second-order selection was visually demonstrated by the localization of the 1 7 Serra Finch territories closely associated with watercourses (Fig. 1). We also found that wet scrubland and riparian wood¬ land habitats delineated establishment of the N JinnBu p\ rr <%) in relali°n ,0 locations of lhe sampled Pairs of Serra Finches in Serra do Cip4 National Park. Braz.l, and the habitat composition within each 95<7r kernel territory, area. Habilal % locations Candeial 16.28 Dry grassland 6.98 Wet grassland 0.00 Rocky outcrop 0.00 Dry scrubland 25.58 Wet scrubland 18.60 Riparian woodland 32.56 % territory Candeial 3 34 Dry grassland 38.77 Wet grassland 130 Rocky outcrop Q.QQ Dry scrubland 27.24 Wet scrubland 19.56 Riparian woodland 9.«o 0.00 0.00 3.51 8.77 0.00 50.88 36.84 0.00 42.32 10.54 5.30 0.00 35.54 6.30 5.77 3.85 3.85 78.85 0.00 7.69 0.00 1.83 30.73 6.41 53.94 0.00 7.09 0.00 0.(K) 0.00 0.00 75.00 0.00 9.09 15.91 0.00 26.46 1.41 62.55 0.00 7.09 2.49 14.29 0.00 7.14 28.57 14.29 21.43 14.29 1.03 62.93 10.98 3.46 4.44 13.57 3.60 0.00 0.00 0.00 69.23 0.00 15.38 15.38 0.00 67.31 5.47 22.78 0.00 2.67 1.78 6.67 0.00 6.67 33.33 3.33 23.33 26.67 0.72 56.73 8.24 14.56 0.61 7.54 11.60 16.67 0.00 0.00 16.67 25.00 16.67 25.00 1.27 73.22 4.74 1.77 4.88 4.21 9.92 0.00 0.00 6.25 25.00 0.00 37.50 31.25 0.00 51.84 18.95 7.69 0.00 16.90 4.62 Freitas and Rodrigues • TERRITORY SIZE AND HABITAT USE OF SERRA FINCH 63 TABLE 4. Habitat ranking matrices for nine Serra Finch pairs in Sena do Cipd National Park, derived from log-ratio differences based on compositional analysis. Triple sign represents significant deviation from random at /’ < 0.05. Second- aider habitat selection compared proportion of habitat used within 05% kernel territories with proportion ot total available habitat in the study area, and third-order compared the proportion of locations for each animal in each habitat type with the piuportionofeach habitat type within the finches's 95% kernel territories. Rank from the least to the most preferred habitat. Candeial Rocky outcrop Dry scrubland Wet scrubland Woodland Second-order Candeial - + - — Rocky outcrop + +■ — Dry scrubland - - — Wet scrubland + + +++ + Woodland + + +++ — Third-order Candeial +++ +++ +++ + Rocky outcrop — + + Dry scrubland — - + Wet scrubland — - - — Woodland - + + +++ boundaries of the pair’s territory. This preference may be the result of higher concentration of fruits, mostly Malastomataceae fruits, which are an important resource for the Serra Finch (Hoffmann et al. 2009; GHSF, pers. obs.). Species of Melastomataceae are highly diverse in the compos rupestres (Giullieti el al. 1997) and produce fleshy "mithucoric fruits, which are abundant in riparian woodlands, rocky outcrops, and scrublands, while almost 'H» fleshy fruits are available in grasslands, die distribution of these particular plants closely mmches the pattern of habitat use by Serra inches in our study area. f bird-order selection examines the specific use 01 ,lat)itut components within the home-range area 'Johnson 1980). Our ranking at this spatial scale 'election indicates a preference for candeial. This vegetation is typically on the western slopes ot the Wi do Cipo (Melo-Junior ct al. 2001), and is Patchily distributed in the study area. This habitat dors hoi supply fruits hut does supply arthropods perches. We also detected more preference 1 r riparian woodland than wet scrubland at this %ale suggesting preference for habitals with w°ody elements. This contrasts with previous '"formation about Serra Finch preference for open 'fetation habitats (Mattes and Sick 1985, vasconcelos 2001. Hoffmann et al. 2009). CONSERVATION IMPLICATIONS We identified the most important habitat types ln foe compos rupestres complex for the Serra Finch and described the spatial distribution that should influence development of better long-term conservation strategies. Actions that contribute to reduce habitat quality for the species should be avoided (BirdLife International 2011). The con¬ servation priority should be maintenance ot the most preferred habitats ot the Serra Finch that po¬ ssibly contribute most to species fitness (Garshelis 2000). Preservation of valleys, woodlands and scrublands, and water in the mountains within the species range must be prioritized; especially important is protection and maintenance of riparian woodlands and candeial. Those habitats have been intensely exploited in the region for production of wood and also for oil extraction in the candeial (Ribeiro et al. 2009). The mountain range where the Serra Finch occurs is critically threatened due to increasing land conversion and no effective conservation plans (Jacobi et al. 2007, Martinelli 2007). The actual habitat protection law is under review which may be detrimental to the currently protected areas (Ribeiro and Freitas 2010). One of the most common threats for the compos rupestres complex are annual fires, mainly promoted to benefit livestock. Annual fires tend to reduce the extent of scrublands and woody elements while increasing the dominance of grassland habitats (Coutinho 1990. Moreira 2000). The Alto do Palacio is a protected area and (here have been no recent reports of fires, unlike in neighboring areas (Franca and Ribeiro 64 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124, No. 1. March 2012 2008). We would expect to find a more grassland- dominated landscape, less woody vegetation, and fewer shrubs in areas with annual burning regimes and, consequently, a lower density of Serra Finches. Thus, areas with annual burning need a fire protection program for habitat restoration and conservation. ACKNOWLEDGMENTS We thank the student volunteers who assisted us in the field. We arc grateful to the TCMBiu .Serra do Cipd staff for lodging and support and to Bruno Crepaldi and G, G. Teixeira for help with the figures. We appreciate the important comments and suggestions on this manuscript by C. E. Braun and L. M. Costa, and the efforts of J. A. Mobley in reviewing our work and helping with the English language. We also thank I'undayAo O Boticario de Pmteyfio a Natureza, CNPq (473428/2004 0 and 473809/2008-7) and FAPEMIG (CRA APQ-0434-5.03/07) for supporting this study. M. R. is grateful to CNPq for fellowship support during this study. LITERATURE CITED Aebischer, N. J.. p. a. Robertson, and R. E. Khnwakd. 1993. Compositional analysis of habitat use from animal radio-tracking data. Ecology 74:1313-1325. Alves. M. A. S. 1990. Social system and helping behavior in the White-banded Tanager {Nfolhraupis fascia la). Condor 92:470—474. Anich. N. M„ T. J. Benson, and J. C. Bednarz. 2009. Estimating territory and home-range sizes: do singing locations alone provide an accurate estimate of space use? Auk 126:626-634. Bibby, C.. M. Jones, and S. Marsden. 2000. Expedition field techniques: bird surveys. BirdLjfe International, Cambridge. United Kingdom. BirdLife International. 2011. Species factsheet: Eniber- nagra longicauda. BirdLife International. Cambridge. United Kingdom, http://www.birdlife.org Braz, V. S. 2008. Ecologia e conservayao das aves campestres do bionta ccrrado. Thesis. Universidade de Brasilia. Brasilia, Brazil. Byers. C. R. and R. K. Steinhorst. 1984. Clarification a technique tor analysis ot utilization-availability dat Journal of Wildlife Management 48: 1050- 1053. Catchpole. C. K. and P. J. B. St a ter. 1995. Bird son biological themes and variations. Cambridge Unive sity Press. Cambridge, United Kingdom. CouTiNito, L. M. 1990. Fire in the ecology of the Brazilk Cerrado. Pages 82-105 in Fire in the tropical biol; ecosystem processes and global challenges (J. ( Goldammer, Editor). Springer Verlag. Berlin, German- Domingues, l. a. l. and m. Rodrigues. 2007. Area d uso e aspectos da terri ton alidade de Schistochlcnm itf itapi "its ( I hraupidae) cm scu penodo >iao-repmdutm R Re^'sla Bras,lcira l|e OmitoJogia 15:538-542 ESRI. 1998. Arc View lor Windows. Version 3.2 Environ saa ~ . . • - ~ Faria. L. C. P., L. A. Carrara, and M. Rodrigues. 2007. Sistema territorial e forrageamento do fura-baneirj Hylocryptus rectirosiris (Aves: Fumariidaet. Revista Brasileini de Omitologia 15:395-401 Franca, H and K. T. Ribeiro. 2008. Mapeamento de queimadas no Parque National da Serra doCipoena Area de Proteyao Arnbicntal Morro da Pcdreini, MG 1984-2007. lCMBio, Technical Report. Ministeno do Meio Ambienlc, Brasilia, Brazil. Freitas. G, H S . 20 111 listriria natural de dois fumandeos (Aves: Furnariidae) cndcmicos dos canipos nipestres da porySo sul da Cadeia do Espinhayo. Minas Gerais. Dissertation. Universidade Federal de Minas Gerais. Belo Horizonte, Brazil. Freitas, g. II. S. and M. Rodrigues. 2007. Canirio rabudo Embcrnagra longicauda: uma ave tipica da> montanlias do leste do Brasil. Atualidades Omitnldgi- eus 141:220-221. GarsiieUS, D. L. 2000, Delusions in habitat evaluation: measuring use, selection, and importance. Pages HI- 164 in Research techniques in animal ecology: contro¬ versies and consequences (L. Boitani and T. K. Fuller, Editors). Columbia University Press. New York, PSA, Giuluei i, A. M„ J. R. Pirani. and R. M. ILrli.y 1997. Espinhayo Range region, eastern Brazil. Pages 397- 404 in Centres of plant diversity: a guide and strategy for their conservation (S. D. Davis, V. H. Hcywood.O. Herrera-MncBrydc. J. Villa-Lobos, and A. C. Hamil¬ ton. Editors). Oxford Press. Oxford. United Kingdom. Google Eariii. 2010. Google Inc.. Mountain View, California. USA. Hoffmann. D. 2006. Forrageamento. dicta, area de vida. biologia reprodutiva e sucessso reprodutivo de Poly- stictus xuperciliaris Wied. 1 83 1 (Aves. Tyrannidae) no sudestc do Brasil. Dissertation. Universidade Federal de Mtnas Gerais, Belo Horizonte, Brazil. Hoffmann, D. 201 I Distribuiyao potential e viabilidade dc uma populayuo de Potysticm superciliaris (Aves. Tyrannidae). no sudestc do Brasil. Thesis. Universi¬ dade Federal de Minas Gerais, Belo Horizonte. Brazil Hoffmann. D.. L. E. Lopes, and M. F. Vasconcelds. 2009. Natural history notes on the Pale-throated Serra- finch (Embcrnagra longicauda) in eastern Brazil Omitologia Neotropical 20:597-607. Jacobi, C. M.. F. F. Carmo, R. C. Vincent, and J R Steh.mann. 2007. Plant communities on ironstone outcrops: a diverse and endangered Brazilian ecosys¬ tem. Biodiversity and Conservation 16:2 185— 2200 Johnson. D. H. 1980. The comparison of usage and availability measurements for evaluating resource preference. Ecology 61:65-71. Johnson, M. D. 2007. Measuring habitat quality: a review. Condor 109:489-504. KERNQHAN, B J , R. A. GlTZEN, AND J. J. MlLLSPACGH. 2001. Analysis of animal space use and movements. Pages 115-166 Radio tracking and animal popula¬ tions (J. J. Millspaugh and J. M. Marzluff, Editors). Academic Press, San Diego. California, USA. Laver, P. N. and M. J. Kelly. 2008. A critical review ot home range studies. Journal of Wildlife Management 72:290-298. Freitas and Rodrigues • TERRITORY SIZE AND HABITAT USE OF SERRA FINCH 65 L£BAN, F. A. 1999. Resource selection for Windows. Version 1.0. University of Idaho. Moscow. USA. http://www.cnrhome.lridaho.edu/fishwild/Garton/tools I.EBAN, F. A.. M. J. WISDOM. E. O. CiARTON, B. K. Johnson, and J. G. Kie. 2001, Effect of sample size on the perfonnance of resource selection analyses, Pages 291-307 in Radio tracking and animal popula¬ tions (J. J. Millspaugh and J. M. Mar/luff. Editors). Academic Press, San Diego. California. USA. Lopes. L. E. and M. A Marini. 2006. Home range and habitat use by Suiriri uflints and Suiriri isteronun (Aves: Tyrannidae) in the central Brazilian Ccrrndo. Studies on Neotropical Fauna and Environment 4 1 :87 91 Makarieva. A. M., V. G. Gorshkov, and B L. Li. 2005. Why do population density and inverse home range scale differently with body size'.’ Ecological Complex¬ ity 2:259-271. Marti NELLI, G. 2007. Mountain biodiversity in Brazil. Revista Brasileira de Botanica 30:587 597 M Afros. G. T. and H. Sick. 1985. Sobre a distribuiyao e a ecologia de duas especies cripticas: Embemagra longicauda Strickland, 1 844. e Embemagra platensis 'Gnielin. 1789). Emberizidae. Aves. Revista Brasileira de Biologia 43:201-206. MeLLO-JCNIOR. T. A.. M. F. V ASCONCELOS, Cl. W. Fernandes, and M. A. Marini. 2001. Bird species distribution and conservation in Serru do Cipo. Minas Gerais. Brazil. Bird Conservation International 11:189-204. Mohr, C. O, 1947. Tabic of equivalent populations of North American small mammals. American Midland Naturalist 37:223-249. Moreira, A. G. 2000. Effects of fire protection on savanna structure in central Brazil, Journal of Biogeography 27:1021-1029. Nnu. c. w„ C. R. Bvers, and J. M. Perk. 1974. A technique for analysis of utilisation-availability data. Journal of Wildlife Management 38:541-545. OBrien, C. E. 1968. Rediscovery of Embemagra long- icauda Sirickland. Auk 85:323. PrrELKA F. A. 1959. Numbers, breeding schedule, and territoriality in Pectoral Sandpipers of northern Alaska. Condor 61:233-264. R- A. 2000. Animal home ranges and territories J»d home range estimators. Pages 65-1 10 in Research techniques in animal ecology: controversies and consequences (L. Boilani and T. K. Fuller. Editors). Columbia University Press, New York. USA. Rmt'fto. B. A.. M. F. Goulart. and M. A Marini. 2002. Aspectos da territorialidade dc Knipotegus lophotes 'Tyrannidae, Fluvicolinae) em seu perfodo reprodu- >ivo. Ararajuba 10:231-235. R|bhiro. K. T. and L. Freitas. 2010. Impactos potenciais das alterayoes no Cddigo Florestal sobre a vegetayao de campos rupestres e campos de altitude. Biota Neotropica 10:239-246. Ribeiro, K. T . J. S, Nascimknto, .1. A. Madeira, and L. C. Ribeiro. 2009. Survey ot the boundaries of the Atlantic Forest in the Serra do Cipti. Minas Gerais, Brazil, aiming at a better understanding and protection of a strongly threatened vegetation mosaic. Naturcza e Conservayao 7:30-49. Rodgers, a. R. and A. P. Carr. 1998. 1IRE: the home range extension for ArcVicw Version 1.1. Ontario Ministry of Natural Resources. Ottawa. Ontario, Canada. Rodrigues, M. and L. A. Carrara. 2004. Co-operative breeding in the Rufous-fronted Thomhird Phacello- dumtts rufif runs: a neotropical ovenbird. Ibis 146:35 1— 354. Rodrigues, M., L. M. Costa, G. H S. Freitas, M. C. Santos, and D. F. Dias, 2009. Ninhos e ovos de Emherizoides Iwrbicola. Emherizoides ypiranganus e Embemagra longicauda (Passeriformes: Emberizidae) no Parque National da Serra do Cipo. Minas Gerais. Brasil. Revista Brasileira de Ornitologia 17:155-160. Rodrigues. M.. G. H. S Freitas. L. M. Costa. D. F. Dias. M. L. M. Varela, and L. C. Rodrigues. 2011. Avifauna. Alto do Paliicio. SciTa do Cipo National Park, slate of Minas Gerais, southeastern Brazil. Check List 7:151-161. Seaman, D. F. and R. A. Powell. 1996. An evaluation of the accuracy of kernel density estimators for home range analysis. Fcology 77:2075-2085. Seaman, D. E., .1. J. Millspaugh. B. J. Kernohan, G. C. Brundige. K. .1. Ralpekk. and R A. Gitzen. 1999. Effects of sample size on kernel home-range estimates. Journal of Wildlife Management 63:739-747. Silva, J. F. 2008. Densidade e lamanho populacional de aves endcmicas c aineayadas dentro da IBA (Important Bird Area) MG06. Dissertation. Universidade de Brasilia, Brasilia. Brazil. Stotz, D. F„ J. W. Fitzpatrick. T. a. Parker III, and D. K. Moskovits 1996. Neotropical birds: ecology and conservation. Chicago LIniversity Progs, Chicago, Illinois. USA. Stutchbury. B. .1. M. and E. S. Morton. 2001. Behavioral ecology of tropical birds. Academic Press, San Diego. California, USA. V ASCONCELOS, M. F. 2001. Pale- throated Serra-finch Embemagra longicauda. Cotinga 16:1 1 0-1 12. Vasconcelos. M. F. 2008. Mountaintop endemism in eastern Brazil: why some bird species from campos rupestres of ihe Espinhayo Range are not endemic to the Cerrado region? Revista Brasileira de Ornitologia 16:348-362. Worton, B. J. 1989. Kernel methods for estimating the utilization distribution in home range studies. Ecology 70:164-168. The Wilson Journal of Ornithology 124(l):66-72, 2012 FORAGING OVER SARGASSUM BY WESTERN NORTH ATLANTIC SEABIRDS MARY L. MOSER'-24 AND DAVID S. LEE' ABSTRACT. — Drifting reefs of Sargassum (a brown alga) are used by a variety of pelagic seabirds in the western Atlantic Ocean. We examined gut contents from 964 individuals of 39 seabird species collected 5 to 60 km off the coast of North Carolina for evidence of Sargassum use. Sargassum pieces or Sargassum- associated prey were found in nine of 10 Procellariiformes species and less frequently among Charadriiformes ( 1 2 of 25 species). No Surgassum-associaied prey was found in Pclecani formes examined, but observational data indicated that Atlantic tropiebirds ( Phocthon Upturns 25% and high volumes of Sargassum -associated prey Audubon's Shearwater ( Puffinus Utemmieri). Royal Tent ( Thalttsseus muximus). Bridled Tern (Onychoprion anaedirtu. n and Red-necked Phalarope ( Pltalaropus lobatus). Seven species fed in Sargassum to a lesser extent, and nine species had ingested Sargassum pieces, but contained no Sargassum-ussoc’e.Hcd prey. It is likely that other seabird species foray; regularly over Sargassum. as our conclusions are based on relatively small sample sizes taken during random sampling in the open ocean, Our conservative analysis and extensive observational data indicate the Sargassum community is critical for leeding lor some western North Atlantic seabirds. Degradation of Sargassum habitats by oil development, harvest, and or ocean acidification would undoubtedly have negative effects on Illness ol these birds Received 22 March 2011 Accepted 7 October 2011. Consolidated reefs of floating pelagic brown algae of the genus Sargassum are important and recurring features of tropical and sub-tropical marine environments. Holopelagic S. natans and S.fluitans support a diverse and abundant fish and invertebrate fauna in the western Atlantic Ocean (Fine 1970, Settle 1993, Casazza and Ross 2008). Recent remote sensing data indicates Sargassum reefs originate in the northwestern Gulf of Mexico in June each year and are advected into the western Atlantic in early summer by ihe Loop Current, moving northward with (lie Gulf Stream (Gower and King 201 1 ). The floating weed moves to the south and west in fall and winter, becoming less buoyant with age. The circulation of Sargas¬ sum is consistent among years and is driven by predictable currents and trade winds (Gower and King 201 1). Floating Sargassum can be extensive, yet ephemeral habitat for seabirds. Airborne imagery indicated that drift lines of the algae extend for continuous lengths ot at least 5 km and primarily consist ot 20-80 nr’ reefs of Sargassum (Marmor- ino et al. 2011). Satellite images indicate ' Zoology Department. North Carolina State Universi Raleigh, NC 27695. USA. u ,CUr?w address; Northwcst Fisheries Science Cem National Marme Fisheries Service. 272 5 Montlake Bou vard East. Seattle, WA 98112. USA, 28336.euSA°iSe R ° B°X 7,,K2' White Lake. J Corresponding author; e-mail: mary.moser@noaa.gov Sargassum slicks can be even larger, ranging from 100 to 1,000 m in width and up to hundreds of kilometers in length (Gower et al. 2006). However, consolidated drift lines of Sargassum off the coast of Florida start to disintegrate as wind speeds exceed 5 m/sec (Marmorino et al 201 1 ). The amount of pelagic Sargassum in the North Atlantic was estimated at 0.54 metric ions; km’ in the Gulf Stream and 0.02 metric lons/knr over ihe Continental Shelf, for a combined standing crop of >50,000 metric tons off the Carolina*; (Howard and Menzies 1969). Gowei and King (2011) estimated the wet weight of Sargassum in the Atlantic has regularly exceeded 1 .8 million metric tons during the past decade and that even greater amounts can occur in the Gull ol Mexico. Thus. Sargassum reefs are important feeding stations and possible roosting sites lot pelagic seabirds (Haney 19S6). Studies to date have used observations of seabird behavior around Sargassum reefs to reach conclusions about why seabirds are attracted to this habitat (Haney 1985. 1986). We examined the gut contents of 39 species of pelagic seabirds for evidence of foraging over Sargassum. Percent frequency ol occurrence, numerical abundance and volume of Sargassum-assoclmed prey weft- used to ascertain the relative extent of Sargassum foraging exhibited by the species sampled. These data were supported by extensive visual observa- lion ol marine birds feeding in pelagic habitats off the coast of North Carolina. Our objectives were 66 Moser and Lee • SARGASSUM REEFS AND SEABIRD FORAGING 67 lo: (1) identify seabird species that rely most heavily on pelagic Sargassum for feeding, and (2) document prey items most frequently targeted by these birds. METHODS Gut content analysis from 964 individual seabirds of 39 species collected 5 to 60 km off the North Carolina coast was conducted as described in Moser and Lee (1992). In addition, visual observations of seabird foraging were made from a vessel during 23 1 day trips (averaging 1 .20 1 seabird observations and 25.2 species/trip). Birds were collected during all seasons between 1975 and 1989. although fewer sampling and observa¬ tion trips were made in winter. Observations and collections occurred over a wide geographic area in ;tn attempt to census inshore coastal waters, the inner and outer Continental Shelf, and deeper waters over the Shelf's edge. Lee and Socii (1998) mapped the areas surveyed by month. Documentation of Sargassum use was not the original focus of seabird collections or observa¬ tions. The birds used in this study were collected opportunistically during their entire period of occurrence in North Carolina waters to obtain data on heavy metal accumulation, plastic ingestion, a£e and sex ratios, body temperature, parasite load, moll sequence, behavior, and ecology (e.g., Moser and Lee 1992, Lee 1995, Lee and Haney !'Wii. Foraging flocks quickly dispersed when approached by our survey boats and birds foraging over Sargassum were not targeted, nor were they particularly easier to collect. Sargassum is typically found in the vicinity of the outer Continental Shelf along the western wall of the rj"lf Stream and. to a lesser extent, in wind rows w'thin the Stream. Only 40% of the surveys were near the Shelf edge where Sargassum typically "Ccun., and the alga was frequently not in the '■"mediate vicinity of our survey sites. Contents from the stomach and gizzard ol each b|rd were combined, and birds with empty upper digestive tracts were excluded from the analysis. Percent frequency of occurrence of Sargassum ''eaves or bladders) and Sargassum- associated Luna was calculated for each seabird species '"umber of birds with prey ‘A' divided by the ,0tol number of birds X 100). Sargassum associates were defined as those species (fish, chistaceans, and gastropods) that reside in Sar- Wmm during the life history phase when '"gested (following Dooley 1972, Settle 1993). Unidentifiable prey items were assumed not to be Sargassum associates, and the mean percent volume of Sargassum- associated prey in the digestive tract was used as a direct measure of the relative importance of Sargassum for foraging within species. Thus, our estimates of Sargassum use are conservative. RESULTS Gnt contents of birds from three Orders, five families, 16 genera, and 39 species were analyzed. Twenty-one species (53.8%) had ingested Sar¬ gassum pieces or Sargassum-ussociated prey (Table l). Birds were classified as Sargassum specialists (species that had >25% occurrence of Sargassum- associated prey). Sargassum users (those with up to 25% occurrence of Sargassum or associated prey), and Sargassum incidentals (species (hat contained only pieces of Sargassum and no associated prey ). We regarded the presence of alga in digestive tracts as evidence of foraging associated with pelagic Sargassum. Its presence among gut contents that lacked any identifiable Sa rgo.v.vHm-associated prey was probably a result of the bird’s inability to rapidly digest the alga. Evidence of Sargassum foraging was found in most Proccl I arii formes (9 of 10 species) and less frequently in Charadriiformcs ( 1 2 of 25 species). It is possible that Sooty Tern ( Onychoprlon fuscatus) could be added lo the species that use Sargassum , as four of 1 1 birds sampled contained Hying fish (Exocoetidae). There was equivocal evidence from digestive tract analysis for Sargassum use by two of the four Pelecani formes. Relatively few indi¬ viduals of these species were collected. Our extensive visual observations of the pelagic members of this family indicated they forage over algal mats. Moreover, two of five White-tailed Tropicbirds ( Phaethan Icpturus) and two of three Red-billed Tropicbirds (P. aethereus) ingested flying fish. Species that had no Sargassum or identifiable Sargassum- associated prey in their digestive tracts included: Band-rumped Storm-Petrel (Oceano- droma castro , n = 12), White-tailed Tropicbird (n - 5). Red-billed Tropicbird (n = 3), Northern Gan net (Morns bassanus. n = 5), Double-crested Cormorant (Phalacrocorax auritus. n — I), Parasitic Jaeger (Stercorarius parasiticus, n = 4), Great Skua (S. skua, n = 1). Great Black-hacked Gull ( Lartis marinas, n = I). Herring Gull (L. argenlatus , n = 2), Ring-billed Gull (L dektwar- ensis, n = 2), Arctic Tern ( Sterna paradisaea . 68 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 n = 2), Caspian Tern ( Hydroprogne caspia, n = 2), Forster’s Tem ( S.forsteri , n — 3), Gull-billed Tern (Gelochelidon ni lotion, n — 3), Least Tem (Sterna la cintillarnm. n = 1). Sandwich Tem ( Thalctsseus sandvicensis , n = 8), Sooty Tem (n = II), and Brown Noddy (Anous stolid us. n = 2). Four species of seabirds had frequencies of Sargassum-associated prey >25% and were considered Sargassum specialists (Table I ). The single Sabine’s Gull (Xenia sabini) sampled contained a seahorse (Hippocampus sp.), which is a Sargassutn dweller. These gulls were observed following Sargassutn drift lines during migration but. sample size did not support including this gull in the Sargassutn specialists category. The four specialists contained almost exclusively Sargassutn- associated prey, as evi¬ denced by high volumes of identifiable prey items in their digestive tracts (Table 2). Sargassutn users, birds that contained Sargassutn prey less frequently (7 species), also Contained high volumes ot Sargownw-associated prey. Most birds feeding in Sargassutn contained small Sargassutn- associated fishes (Table 2). The only exceptions were the two species of phala- ropes, which had consumed Sargassum shrimp (Latreutes fucorunt) and the Sargassutn- associat¬ ed gastropod. Litiopa inelanosioma. This gastro¬ pod was also found in Cory’s Shearwaters ( Calonectris diomedea). Specialists fed on a minimum of seven Sargassum-dssocialed fish species, and most measurable fish prey were <50 mm standard length (Table 2). However. Royal Terns (Thalasseus maximus) generally consumed slightly larger fish (range^ = 40- 105 mm) than the other birds we examined (range - 6-75 mm). Filefish (Monacanthus sp.) occurred with the highest frequency in Audubon’s Shear¬ waters (Puff inns Iherminieri ), Bridled Terns ( Onychoprion anaeihetus), and Royal Terns, but numerical abundance of filefish was highest only in Audubon's Shearwaters and Royal Terns (Fig. 1A. C). The mean number of filefish per bird was five. Bridled Terns (Fig. IB) had the highest number of tetradontids (mean number of puffers/bi rd = 6) and also consumed large numbers of ostraciids (mean number of trunk- fish/bird = 4). DISCUSSION Gut content analysis identified four Sareasv, spec.ahsts, while visual observations indicat ha. several add.tional species target this habi for feeding. Audubon’s Shearwater. Royal Tem. Bridled Tern, and Red-necked Phalarope iPhahtr- opus lobatus) contained relatively high levels a Sargassutn- associated prey. The single Sabine- Gull examined contained a 5argaw«m-associaied prey item and this species was observed to follov Sargassutn drift lines. Visual observations indi¬ cated Bridled Terns regularly associated with Sargassutn patches and tended to use the mats and associated flotsam as roosting sites (Duncan and Harvard 1980, Haney 1986). Our observational data also indicated that Masked Boobies I Sulo dactylatra) and the two species of tropicbinh target Sargassutn patches while feeding as also reported by Haney et al. ( 1999). Diet analysis underestimated prey from Sargas- sum habitat and excluded some seabird species. For example, the digestive tracts of Bridled Terns in our study contained insects of terrestrial origin, which we did not consider to be Sargassutn associates (5 consumed Lepidoptera, 6 ate Coleoptera. 2 ale Hymenoptera. and 6 contained unidentified in¬ sects). Haney ef al. (1999) reported insects were (lie second most common food item in Bridled Temv These insects may have been resting on Sargassutn mats when ingested, We commonly observed both species of Atlantic tropiebirds feeding around Sargassutn reefs. Flying fish were recovered from their digestive tracts, but this was not direct evidence of Sargassum use. as flying fish regularly occur where Sargassum is absent (Casa/za and Ross 2008). We observed tropiebirds as they plunged near and sometimes under the Sargassum The same was true of Masked Boobies, a species seen infrequently off the Carolinas but usually tn association with Sargassum. Some seabird species may not feed directly over Sargassum , but the alga is critical habitat Ion- certain life stages of their fish prey. For example (lying fish can represent >50% of the total diet ot the two tropiebirds collected off North Carolina ( Lee et al. 1981, Lee and Irvin 1 983 1 and Hying fish are important prey in other parts of their range (Lee and Walsh-McGchee 1998). Flying fish use Sargassum for spawning and rearing, and Sargas- sum is essential habitat for these and other fish species (Casa/za and Ross 2008). Thus. Sargassum contributes indirectly to the fitness of tropiebirds and other seabirds, notably Sooty Terns and Masked Boobies. Sargassum specialists used a variety of foraging modes, including surface-seizing, plunging (aerial diving), pursuit plunging/diving, pattering, and Moser and Lee • SARGASSUM REEFS AND SEABIRD FORAGING 69 x. O'. 8 £ _u , 'a vd 0 >*£ o c — S 0 w i! i § S E 0 0 | A “ g Ci o “ c q u E 2 C3 ■_ " 2 -5 „• a .b ■D f X o 5 o> c ^ c 5 s l < o I 0 x - Z « o ° o H S « s> 8 V £ * 5 a ^ c 2 5 |-S£ -M I 11! O « c i £ | § V .5 CJ " U «§} T3 ■© -5 5 5 c 11 g &&1 o w > ^ 5 - «_ $ c ° § g £ I a 8 3 § C -a « £2 £ 2 I ^ «5 I’ll o *3 frBl iti a- SJ ? U- -J M •j ~ .2 «o u £ a -i * y ca » > <55 H 2 s O' O >/-> o c o o o o c o — — c O q cj — O' o q q o q 00 o o — c Tl- rf f*~. O O tN Ifiooco ONOOO O'f'io — c 5 a. 2 4- £ 5 2 ~ -2 '5 = a. ~ 9- 5 O «o J a J I a p: "3 E 1) *r •© >• o c cc ac 1| 5- d S' > 6 < E = q q ci — q oo o w d 't vd x •t w n — — — o v© fi f'l i/1 e*-. — O' -o f B S' s s -a -o E E H 1: -o £_ " E < §§ -O T3 E E | | ■5. .2 8H v r- *& , a'q g U3 ^ ,1 %■ "? ; to *- — ii- 2 B I i a o 1 ^ || S-tjj i2 XI H p • CO E?Ul ro w O • S ? Ijv or S = x a E s H. = U?S: XT O B O C 2 fld ■ — -fix-, r-, — — r~- c O d (N fd rd xi id O C O vd o - i r, n r--. — m r^, a u oo 1 OJ t>0 3 b g- 22^ S3 «L^. < — " k— 3 Cl) -> ^ oo >:• § S3 3 ? 2 'f J Cv *~ « J ~ - s ? s < o c5 "© "© « >»’e e ?! u o o o ^ 1 E >■! S3 f T > il - 5-? R. -5 co ^ ~ 2 ~ 5 S 1 S ^ — 1s£N S' ^ — .2 s afo g & | I| "5= '5 §j fc I :1 g i ^ '.«v ^ a. _ -3 5 .& g _ ',L-odu^;.C 2 — — o a o ^ -s — S •-> o D. ■ b so - -S .= a >_r-C£roftirO’ Mi ei "» P s nj w _J 03 U O J * I £ 3 ^ ; to 3 -c C a .= f- i ? OjrJrZ^'J: < ?uSic-PorJ ? ; 2 u 0 ' - _r _o v/i/5 a j^Z jooe 70 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 TABLE 2. Frequency of occurrence (%) and size range (standard length in mm) of individual fish taxa ingested by three Sargassum specialists: Audubon’s Shearwater (/i = 48), Bridled Terns (n = 16), and Royal Terns (n = 8). Frequency of occurrence (%) and size range (mm) Audubon's Shearwalcr Bridled Tern Royal Tern Exocoetidae (flying fish) 6.2 (40-60) 6.2 (30) 0 Syngnathidae (pipefish) 0 6.2 (23) 0 Priaccmlhus sp. (bigeye) 4.2 (15) 12.5 (20) 0 Heteropriacamhus cruentatus (glasseye) 2.1 (45) 0 0 Caranx sp. (jack) 14.6 (45-50) 0 12.5 (70) C. hippos (erevalle jack) 2.1 (25) 0 0 Decapterus sp. (scad) 4.2 6.2 0 Trachurus laihami (rough scad) 2.1 (75) 0 0 Stromatcidae (buiterfish) 4.2 (15) 12.5 (30-40) 12.5 Psenes sp. (driftfish) 0 0 12.5 (70-105) Balistidae (triggerfish) 8.3 (15-20) 25.0 (20) 0 Aluterus sp. (filefish) 12.5 (40) 0 12.5 (40) Monacanthus sp. (filefish) 68.8 (15-50) 31.2 (10-22) 62.5 (40) Monacanthus ciliatus (fringed filefish) 2.1 (25) 0 0 Stephanolepis hispidus (planehead filefish) 2.1 (30-52) 0 12.5 (45) Uictophrys sp. (trunkfish) 0 12.5 (6-10) 0 Chilomycterus sp. (biinfish) 0 0 12.5 Sphoeroides sp. (puffer) 2.1 (7-12) 18.7 (10-20) 0 S. maculatus (northern puffer) 0 6.2 (10-12) 0 dipping (following Ashmole 1971). We observed Audubon s Shearwater feeding near the surface, either by shallow diving (1-2 m), surface-seizing, or hydroplaning in and around Sargassum reefs. However, in the Bahamas, this shearwater feeds by pursuit diving during the nesting season with dives averaging 7.6 m (n = 136) to a maximum of 29 m (Mackin 2004). Phalaropes. which prey on aquatic invertebrates in shallow pools in the tundra by surface feeding (Haney 1985), use the same behavior when seizing snails and crusta¬ ceans from Sargassum mats. The spinning beha\ ior associated with phalaropc feeding in freshw. ter habitats (Obst ct al. 1996) was not observed ; sea. Sargass um-associ ated prey taken by sma seabirds were rarely >50 mm. indicating thes birds picked prey from within the alga, as oppose to diving beneath it where larger fish are typical I found (Moser et al. 1998). An advantage c foraging in Sargassum reefs is that piscivorou predators drive prey up into the Sargassum mat? where it is more accessible to the smaller-bodie seabirds (Safina and Burger 1985, Haney 1986). Prey types in digestive tracts provided additions information about seabird feeding. Both frequenc ,°n rCTTL? Hnd numeric«' abundance offish pre: ■n Bridled Terns indicated they select relative.; uncommon members of the Sargassum fish fauna tetradontids (puffers), ostraciids (trunkfishes), stro- mateids (driftfish), and priacanthids (bigeyes) (Dooley 1972, Settle 1993). However, these fishes may occur at the periphery of Sargassum patches, where they are less frequently collected during Sargassum sampling with nets (Casazza and Ross 2008). In contrast. Royal Terns, Audubon’s Shear¬ waters. and Red Phalaropes fed on prey that are dominant members of the Sargassum community- filefishcs. jacks, and Sargassum shrimp (Fine 1 970. Dooley 1972, Settle 1993, Casazza and Ross 2008). Haney (1986) found a significant relationship between bird hotly size and Sargassum patch size We noted that large-bodied Royal Terns contained relatively large prey: but this was likely a function of their feeding mode (plunging) rather than Sargassum patch size. Sargassum foraging w as documented during all months of the year despite the Sargussuin mat structure and attendant fish community changing seasonally and in response to w eather I Moser et al. 1998, Casazza and Ross 2008. Gower and King 2011). Fine (1970) noted that faunal composition in Sargassum collected from tk Gulf Stream and Sargasso Sea was similar, bid that non-colonial macrofauna were more abundant in spring than in full. This may affect the way seabirds use Sargassum habitat. Royal Terns Moser and Lee • SARGASSUM REEFS AND SEABIRD FORAGING 71 I Numerical percentage (numbers of prey item ^ divided by the total number of prey items X 100) ol V«a«unt-associated fishes in the digestive tracts of Chon's Shearwaters (A) (// = 48). Bridled Terns (Bl 16). and Royal Terns (Cj (n = 8). ' ■"ntnute daily from Outer Banks nesting colonies ,IJ forage in Sargassum mats along ihe edge ol ,hc outer continental shelf, a round trip of up to 160 km or more (DSL. unpubl. data). Common lS,fw hirundo) and Black ( Chlidonias niger ) lenis, and Sabine's Gulls seasonally migrate north ^ south, and likely use rows of Sargassum along the western edge of the Gulf Stream and drift lines within the Stream to both orient and feed. Reduction in the Sargassum community would have negative effects on a number of western North Atlantic seabirds, based on digestive tract analysis and at-sea observations, including five tropical species considered to be of conservation concern (Schreiber 2000). Observations of Ber¬ muda Petrels (Pterodronui t allow) and Roseate Terns ( Sterna dougallii) indicated these Endan¬ gered Species also use Sargassum to forage. Sargassum use by seabirds in the Pacific and Indian oceans is unknown, but it is likely that many of the same species and their ecological counterparts exploit Sargassum reefs in those oceans as well. Commercial harvest threatens to reduce the standing crop of Sargassum in the western North Atlantic (Settle 1997). and there are possible negative impacts to Sargassum from oil and gas exploration on the outer Continental Shelf off the coast of North Carolina (Lee 1999). Seabird surveys in the Gulf of Mexico shortly after the Deepwater Horizon oil spill (April 2010) indicated Sargassum habitat was damaged by this event (J. C. Haney and DSL. unpubl. data). Global climate change and attendant ocean acidification may also affect Sargassum (Porzio et al. 201 1). Significant reduction in the amount or quality of Sargassum habitat could reduce seabird abun¬ dance, influence marine distribution, alter season¬ al movements, and/or jeopardize the birds physiological condition. ACKNOWLEDGMENTS Wc thank S. W. Ross Cor help with fish identification and Cor reviewing an early version ol this manuscript. H. J. Porter aided in gastropod identification. .1. M. Butzerin, T P. Good. J. C. Haney. W. A. Mackin. R. L. Pitman, and an anonymous reviewer provided helpful reviews of this manuscript. Funding Cor this study was provided in part by Ihe National Marine Fisheries Service, U.S. Fish and Wildlife Service. U.S. Army Corps of Engineers, and U.S. Department of the Navy. LITERATURE CITED Asiimole. N. P. 1971. Seabird ecology and the marine environment. Pages 223-286 in Avian Biology (D. S. Famer and J. R. King. Editors). Volume J. Academic Press. New York, USA. Casazza. T. L. and S. W. Ross. 2008. Fishes associated with pelagic Sargassum and open water Jacking Sargassum in the Gulf Stream off North Carolina. Fishery Bulletin 106:348-363. Dooley. J. K. 1972. Fishes associated with the pelagic Sargassum complex, with a discussion of the Sargus- 72 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 sum community. Contributions in Marine Science 16: 1-31. Duncan. C. D. and P. W. Harvard. 1980. Pelagic birds of the nonhem Gulf of Mexico. American Birds 34:122- 132. Fine, M. L. 1970. Faunal variation on pelagic Sargassum. Manne Biology 7:112-122. Gower, ). and S. Kino. 2011. Distribution of floating Sargassum in the Gulf of Mexico and Atlantic Ocean mapped using MER1S. International Journal of Remote Sensing 32:1917-1929. Gower. J„ C. Flu. G. Borstad, and S. King. 2006. Ocean color satellites show extensive lines of floating Sargassum in the Gulf of Mexico. IEEE Transac¬ tions on Geoscience and Remote Sensing 44:3619 3625. Haney. J. C. 1985. Marine distribution, seasonal abun¬ dance, and ecology of phalaropes in the Georgia Embayment. The Oriole 50:21-31. Haney. J. C. 1986. Seabird patchiness in tropical oceanic waters: the influence of Sargassum "reefs". Auk 103:141-151. Haney. J. C.. D. S. Lee, and R. D. Morris. 1999. Bridled Tern ( Sterna anueihetus). The birds of North America. Number 468. Howard, K. L. and R. J. MENZIES. 1969. Distribution and production of Sargassum in the waters off the Carolina coast. Botanica Marina 1 1-12:244-254. Lee. D. S. 1986. Seasonal distribution of marine birds in North Carolina waters. 1975-1986. American Birds 40:409-412. Lee. D. S. 1995. Marine birds off the coast of North Carolina. Chat 59:1 13-171. Lee. D. S. 1999. Pelagic Seabirds and the proposed exploitation for fossil fuels off North Carolina: a test for conservation efforts of a vulnerable international resource. Journal of the Elisha Mitchell Scientific Society 115:294-315. Lee, D. S. and J. C. Haney. 1996. Manx Shearwater (Puffin us puffinus), The birds of North America. Number 257. Lee. D. S. and E. W. Irvin. 1983. Tropicbirds in the Carolinas: status and period of occurrence of two tropical pelagic species. Chat 47:1-13. Lee, D. S. and M. C. SOCCI. 1998. Potential effects of oil spills on seabirds and selected other oceanic verte¬ brates off the North Carolina coast. Occasional Papers North Carolina Biological Survey 1998-1. Lee, D. S. and M. Walsh-McGehee. 1998. White-tailed Tropicbird (Phaethon lepturus ). The birds of North America. Number 353. Lee, D. s . D. B. Wingate, and H. w. Kale II. 1981. Records of tropicbirds in the North Atlantic and upper Gulf of Mexico, with comments on field identification American Birds 35:887-890. Mackin. W. A. 2004. Communication and breeding behavior of Audubon's Shearwaters. Dissertation University of North Carolina. Chapel Hill. USA. Marmorino. G. O.. W. D. Miller, G. B. Smith. andJ.H Bowt.ES. 2011. Airborne imagery of a disintegrating Sargassum drift line. Deep Sea Research I 58:316-32!. Moser. M. L. and D. S. Lee. 1992. A fourteen-year survey of plastic ingestion hy western North Atlantic scahint.. Colonial Waterhirds 15:83-94. Moser. M. L.. P. J. Auster, and J. B. Bichy. 1998. Effects of mat morphology on large Surgassum-msochlei fishes: observations from a remotely operated vehicle (ROV) and free-floating video camcorders Environ¬ mental Biology of Fishes 51:391-398. ons i . B. s.. w Hamner, e. Wolanski. M. Hammer, ami M. RuBEGA. 1996. Kinematics of phalarope spinning Nature 384:121. Por/.io. I ... M. C. Buia, and J. M. Hall-Spencer. 2011 Effects of ocean acidification on macroalgal commu¬ nities. Journal of Experimental Marine Biology and Ecology 400:278-287. Safina. C. and J. Burger. 1985. Common Tern foraging: seasonal trends in prey fish densities and competition with hluefish. Ecology 66:1457-1463. ScTlREtBER. E. A. 2000. Action plans for conservation of West Indian seabirds. Pages 182-191 in Status and conservation of West Indian seabirds (E. A. Schriber and D. S. Lee, Editors). Caribbean Ornithology Society Special Publication Number 1 Settle. L. R. 1993. Spatial and temporal variability if* d- distribution and abundance of larval and juvenile fishes associated with pelagic Sargassum. Thesis University of North Carolina. Wilmington. USA Settle, L. R. 1997. Commercial harvest of pelagic Sargassum: a summary of landings since June 1995 Pelagic habitat Sargassum and water column. South Atlantic Fishery Management Council essential fch habitat workshop Number 9: 7-8 October 1997. South Atlantic Fishery Management Council. Charleston. South Carolina. USA. The Wilson Journal of Ornithology 124(1 ):73-8(), 2012 MALE COMMON LOONS SIGNAL GREATER AGGRESSIVE MOTIVATION BY LENGTHENING TERRITORIAL YODELS JOHN N. MAGER III.' 4 CHARLES WALCOTT.2 AND WALTER H. PIPER' ABSTRACT— We examined two critical predictions of the hypothesis that male Common Loons (Gavin irnmer) communicate greater aggressive motivation by increasing the number ol repeat syllables within their territorial yodels. We observed (from >3.500 hrs of field observations of 58 males) the probability that territorial interactions escalated from territorial flyovers by intraders to stereotyped 'social gatherings' to escalated tights between residents and intruders was positively correlated to the number of repeat syllables given by individually-banded males. Males yodcling during these escalated contests often assumed the upright 'vulture' posture rather than the usual 'crouch' posture, reflecting an escalated aggressive motivational state. Territorial pairs responded sooner and with more threat and alarm vocalizations to playback yodels that contained more repeat phrases. This reflected a greater willingness to attack by residents to perceived intrusions by males of higher aggressive motivational slate. Our study demonstrates the ability ol loons to communicate greater aggressive motivation by lengthening acoustic territorial threat signals, which not only may be important toi conveying imminent attack, but may also reflect important tactics for individuals of poorer lighting ability to deter territorial es ictions. Our results also raise questions regarding what receiver-dependent and receiver-independent selective factors are responsible for maintaining signal honesty in this non-osdne bird. Received M January 2011. Accepted 15 July 2011. Most threat signals communicate information regarding an individual's inherent, or condition- dependent fighting ability (Parker 1974) and/or willingness to attack (or aggressive motivation) (Maynard Smith 1982, Bradbury and Vchrencatnp 1998. Hurd and Enquist 2001). Features that communicate fighting ability often reflect stable physical attributes, like physical size, that predict success in aggressive encounters (Parker 1974. Archer 1988). An animal’s willingness to attack is Men influenced by ephemeral factors such as health and motivational state of both the animal and its competitors (Maynard Smith 1982, Hurd ai>d Enquist 2001), Many birds benefit from communicating varying levels of aggressive moti¬ vation wuthin territorial signals to avoid conflicts ‘hat consume energy and can cause serious injury. B'r One University Drive. Orange, CA 92866. USA, Corresponding author: e-mail: j-mager@onu.edu Male Common Loons (Gavia irnmer) defend all¬ purpose territories on freshwater lakes by aggres¬ sive threat vocalizations called yodels (Sjolander and Argen 1972: Rummel and Goetzinger 1975. 1978). Yodels are given only by male Common Loons, and are considered to be territorial threat signals because males aggressively approach and yodel at eonspecific territorial intruders (Vogel 1995; J. N. Mager. unpubl. data). Structurally, yodels consist of a 3-4 note introductory phrase followed by a strophe of two-syllable repeat phrases (Fig. I). Most frequency elements of a yodel exhibit low intra-individual variability (Barklow 1979; Vogel 1995; Walcott et al. 1999; Mager et al. 2007a, b) and each male can lengthen yodels by increasing the number of repeat phrases. Barklow (1979) observed males added more repeat phrases to their yodels when intruders wandered deeply into breeding territo¬ ries, and suggested that because intrusions pose a greater threat to resident males, longer yodels might communicate a greater willingness to attack. This hypothesis has been generally accept¬ ed. but has been supported by few. and only anecdotal, data. For example, males add repeat phrases to yodels when territory quality (and resource value) is enhanced (Mager et al. 2007b), which may indicate they defend higher quality territories more aggressively. There has been no published study to date that has examined critical predictions of this long-standing hypothesis. We specifically tested Bark low's hypothesis that yodels containing more repeat phrases reflect a greater motivational state by examining two 73 74 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. 1. March 2012 FIG. 1 . Spectrogram above) indicating change in frequency (in kHz.) over time (sec) and waveform (below) indicating ? "!L 'ff (MPa.) over time (sec) of a typical yodel of the Common Loon. Yodels consist of two fundamental varv'thp “ t "“f ,ntroduc'ory Phrase’ followed *>.v a scries of two-syllable repeat phrases in which each individual can vary the number ol repeat phrases for each of the yodels they produce (this example has 5 repeat phrases). critical predictions: (I) males should give longer yodels during social situations in which (he probability of attack is greatest, and (2) territorial pairs should respond more aggressively to yodels that vary only in number of repeat phrases. METHODS Stmly Site and Resident Population.— W( conducted this study from 2002 to 2007 or ~100 study lakes in Oneida County. Wisconsin USA, northwest of Rhinelander (45° 42' N 89- 37' W). Loons are genetically monogamous and exhibit high territory and male fidelity (indi¬ viduals return to the same territories for 3-7 years on average; Piper et al. 1997a). Territorial intrusions occur frequently between April and July (Mageranc Walcott 2007) and occur most often between 043C and 1 100 hrs each day (Mager 1995 ). Assessing the Effect of Social Context on Yodel Duration.— We assessed the impact of varying number of repeat phrases on behavior ol signalers and receivers by considering whether longer yodels were given during five stages of the intrusion process in which the probability of resulted* f" ** pr°porlion °‘‘ situations that Pre.l ^ conspecifics flew over the' \eZory, % when conspeafics landed on the territory bu, remained fr°m the res,dem male, (3) when intruders approached to within 20 m of the resident male bin did not engage in ‘social gatherings' that consist of stereotyped circling on the water's surlace in head-to-tail orientation that may include ‘splash dives’ (Sjdlander and Argen 1972) with pair members, (4) when participants engaged in social gatherings, and (5) when intruders attacked or fought a resident. Daw collected between 2002 and 2004 verified that residents were more likely to attack an intruder (repeated-measures ANOVA F3.246 = 1.654. P < 0.0001; Fig. 2 ). when intrusions transitioned from flyovers to actual landings (paired / = 8.093. df - 1^?- P < 0.0001). from landings to approaches (paired t = 2.178. df = 155. P = 0.0073), and from approaches to social gatherings (paired t = 6.969, df = 155, P < 0.000 1 ; Fig. 2). Observations of Natural Yodels.— We observed 58 pairs of individually-banded loons between I- April and 31 July 2002-2007 between 0430 and 1430 hrs (CDT). We conducted daily M11 behavioral-time samples (Altmann 1974) of 4-6 territorial pairs, using ‘all-occurrences’ recording methods (Martin and Bateson 1993) to count flyovers, the extent of intrusions (e.g.. did pairs approach within 20 m of the intruder, did die residents and intruders engage in 3 social gathering), aggressive behavior (e.g.. surlace chases involving the swimming and flapping of wings of one individual toward another, or actual Mageretal- LOONS SIGNAL HIGH AGGRESSION WITH LONGER YODELS Flyover Distant intruder Close intruder Social gathering FIG. 2. Mean (± SE) probability of attack under different social contexts based upon >3,530 hrs of observation of Common Loons between 2002 and 2004 (/•\4g = 75.448. P < 0.0001, n = 84 males). fights in which individuals actually grab one another by cither the bill or neck and beal one another with their wings, following descriptions in McIntyre (1988) and Piper et al. (2008), and yodels. We recorded the time of day. the yodeler's physical posture (i.e.. ‘crouch’ or ‘vulture’ after Rummel and Goetzinger 1978). the estimated distance between yodeler and perceived receiver, and the number of repeat phrases given when each male yodeled. Playback Experiment.— We assessed how loons responded to yodels having fewer/more repeal syllables by recording vocal responses of local pairs to broadcast yodels that simulated an unfamiliar male’s intrusion into the territory. We ““d a Portadat MDP 500 acoustic recorder HHB. London, UK) equipped with a Sennheiscr MHK-70 shotgun microphone (Old Lyme, CT, ^A) to broadcast yodels from males recorded e«fflier that season at a distance between 15 and ^ m from males in the study area whose territories were >8 km from a focal pair's territory (to reduce social recognition) (Waas lwk Mager et al. 2010). and did not play the ^me yodel to more than one pair (to prevent pscudoreplication ) (Kroodsma 1989). Wc created undistorted yodels containing one. lour, and seven repeat phrases by adding identical repeat phrases t'Jthe same yodel using Canary acoustic soilware lVenion 1.5, Cornell University Bioacoustics ^‘search Program, Ithaca. NY, USA), Thus, each yodel was identical in all acoustic parameters ex«pt for the number of repeat phrases. We playbacks between 2130 and 0230 hrs on three successive nights. We randomized the cider yodels were played on the three nights (e.g., 1 -repeat phrase the first night, 7-repeats the second, and 4-repeats the third night to one pair; 4, 7, I repeat phrases on successive nights to another pair, etc.). Each night, we broadcast yodels through a RadioShack (Fort Worth, TX. USA) 20-Watt amplifier and Super PowerHorn model 40-1 .445 speaker at 90 dB (measured 10 m from the speaker) and rebroadcast the same yodel (having the same number oi repeat phrases) at 5 and 10 min after the first. We recorded the time and type of all vocal responses during the 15-nun period following the first playback using the same recorder and microphone from which we recorded the playback yodels, We quantified the number of tremolo, wail, and yodel responses from focal pairs in addition to the latency before first vocalization. We interpreted the number of tremolos to reflect the extent pairs were threatened by the yodel, and the number of wa.l responses to reflect the pair’s level of alarm and interest to contact following Barklow (1979) and McIntyre (1988). We used repeated-measures ANOVA and associated post-hoc tests to investigate differential responses hy pairs to the three classes ol playback. We used the PROC MIXED procedure (SAS after Singer 1998, Johnson 2002) to construct linear growth models to examine whether responses given could be attributed to more repeat phrases while controlling for variability among pairs. We accepted significance at a Bonferroni -corrected a of 0.05 for all statistical tests. RESULTS We recorded conspecific intrusions on 57 territories and the yodeling responses of 58 banded males defending those territories (x ± SE number 76 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. I. March 2012 of Fclch' oerS!!=n„“,mbe; m T?" ^ C°mmon L°°ns * ”cial “»««■ Gray bats reprcSen, number or crouch yodels per event and black bars represent number of ‘vulture' yodels per event. of seasons an individual was observed = 3.7 ± 0.2 median = 3 seasons) during 3,900 hrs of observation covered by our 6-year study. The number of prospecting loons flying over territories varied significantly with each territory (F5M3I = 1.718, P = 0.0051), but the frequency of distant intrusions (F56, 15 , = 1.505, P = 0.0268), ap¬ proaches (F5615| = 0.862, P = 0.7349) social gatherings (FS6,J51 = 1.316, P = 0.0975), and aggressive chases/fights (F5()il5, = 1.202, P = 0.1914) did not vary with territory. Intrusions that involved the close approach of the intruder without a social gathering occurred rarely; consequently, we report mean yodel ing responses by males to these types of intrusions, but omit them from repeated-measures analyses of vodeling rates and number of repeat phrases given with stage of territorial intrusion. None of the contests resulted in eviction of territorial males. we observed 1,495 yodels from 58 banded m (mean ± SD; 13 ± I yodels recorded/individ year, range = 1-51). Forty-eighl of the 58 m; yodeled in al least two of the five social com, between 2002 and 2007; few gave yodels in all f Pooling all yodels from all males, males yotk primarily at conspoeiRcn (87.5% of all yodels) also yodeled al humans ( 10.8%). and Bald Eat hetT“ '*mac‘ph“lu-,> d-7%). We obser siom T'S i°',ni"in8 infn-‘‘'ucnl in, territorial in^o, 48 0^56 1 years). n,e frequency (# yodelsfoccurrenee) ma yodeled at conspecifics was related to the extent of intrusion (Friedman’s ANOVA f = 25.394. n = males, df = 3, P < 0.0001; Fig. 3). Males gave significantly more yodels when intruders landed on the territory than when intruders flew over the territory (Wilcoxon Z = 2.99. P = 0.0028). and gave progressively fewer yodels per social gathering (Wilcoxon Z = 2.92. P = 0.0035). However, male vodeling rate did not decrease as social gatherings escalated into chases or fights (Wilcoxon Z = 0.62. P = 0.54). Close intrusions that did not lead to social gatherings rarely occurred, but the likelihood that a resident nude would yodel was lowest (14.3% of all such occurrences) during this situation, implying some signal, or assessment of lower threat, associated with the intruding loon. Males gave most calls as ‘crouch’ yodels (94% °l given), but only gave crouch yodels at flyovers, distant intruders, and close intruders that did not escalate into social gatherings. Males gave more 'vulture-posture’ yodels when territorial contests escalated; 1 0. 1 % of all calls given during social gatherings were crouch yodels, but 678f of all yodels given during physical chases, and/or fights were vulture-posture yodels (Fig. 3). Male' (/i ' 25) observed yodel ing in both the crouch and vulture postures during a given vear gave yodels with more repeat phrases when in the vulture- posture (X ± SE = 4.95 ± 0.40) than in the crouch-posture (3.72 ± 0.26; paired l = 3.10. df - 24, P = 0.0049). Intruders rarely yodeled on territories of residents (130 yodels in the 6 years Mager el al. • LOONS SIGNAL HIGH AGGRESSION WITH LONGER YODELS 77 6.0 Flyover Distant intruder Close intruder Social gathering C base Fight FIG. 4. Mean (± SE) number of repeat syllables per yodel given by resident male Common Loons under social contexts associated with greater levels of agression. of study), and tended lo give proportionately more vulture yodels (33/130 or 25.4% of yodels) than residents (84/1.495. or 5.6% of yodels). Males had 3.6 ± 0.1 repeat phrases/yodel (range = 0.0-20.0). although the number of repeat phrases each male produced was a function of the number of yodels heard (y = 3.320 log x + 1.064; r = 0.613; P < 0.0001). Most yodels (S6.l%) contained between two and five repeat phrases (mode = 3). There was individual variation in the number of repeat phrases given hv males per crouch posture yodel (ANOVA ^7.129 = 3.270. P < 0.0001 ). but not per vulture Posture (ANOVA F2S, 12 = 1.431. P = 0.26) .'"del. not controlling for the different contexts of territorial intrusion. Males had 42.9% more repeat phrases per yodel as intrusions escalated from flyovers and landings to wia| gatherings and physical attacks (Fig. 4). Males ^ lunger yodels when flyovers escalated to landings 'Nred 1 = 3.158, df = 39, P = 0.003 1 ); did not give ’significantly longer yodels when intruders came U|1hin 20 in of the male (paired / = 0.025, dl = P ~ 0.98); and gave longer yodels when close "'tnisions transitioned into social gatherings (paired t = 3.I68. df = 10, P = 0.01), but not when social gatherings transitioned into chases and/or fights (paired t — 0.09, df = 17, P — 0.93). There was a clear difference in how pairs responded vocally to broadcast yodels that differed in the number of repeat phrases. Pairs vocalized over twice as quickly, and gave almost tour times as many tremolos to longer yodels. Resident males gave four times as many yodels to seven-repeat yodels versus one-repeat yodels (Table I ). DISCUSSION Individuals that experience frequent territorial contests often produce graded threat displays to communicate a heightened state of aggressive motivation (van Rhijn 1980, Bradbury and Vehrencamp 1998). Populations experience fre¬ quent conspecific intrusions that escalate into potentially lethal confrontations (Piper et al. 1997b. 2000), and it is not surprising that male Common Loons, as Barklow (1979) proposed, lengthen territorial yodels to communicate greater aggressive motivation. Territorial interactions between residents and intruders among loons proceed through a se- T«U I. Vocal responses of territorial parrs („ - 3S> of Common Loons to playback yodels having one. four, and ieven rePeai phrases. Response One repeal Four repeats Seven repeals Individual growth model Fi n j p bivariate response ^ ^0Cali7fltinnc 9.68 ± 2.96 17.80 ± 5.06 26.76 ± 4.85 6.89 0.0099 * Tremolos 5.92 ± 2.79 14.72 ± 4.91 20.08 ± 4.53 5.26 0.0237 * Wails 3.95 ±1.14 2.44 ± 0.79 4.54 ± 0.91 0.12 0.7288 # Yodels 0.18 ± 0.08 0.36 ±0.14 0.73 ± 0.20 6.51 0.0121 patency before 1st vocalization (sec) 128.20 ± 52.72 131.72 ± 49.12 62.87 ± 17.27 12.37 0.0006 78 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. 1. March 2012 quence of actions that provide more reliable information about fighting ability but entail greater costs consistent with sequential-assess¬ ment models of contests (Enquist and Leimar 1983. 1987). Flyovers provide intruders informa¬ tion about territory quality (Piper ct al. 2006). but if a resident yodels, it also provides the prospect¬ ing loons information about territory occupancy, the resident male's identity (Walcott et al. 1999, 2006; Mager et al. 2010), and fighting ability (Mager et al. 2007a). Flyovers constitute a lesser threat to a resident male versus a close intruder. Males yodel less often (-1 of every 3 flyovers), and always assume the low-risk, low-cost crouch posture. Not only is the “crouch" posture less costly physiologically to assume versus the "vulture" posture, but it also places the signaler in a less vulnerable posture for receiver attack (although no attack could possibly occur during a flyover) and a more effective posture to transmit information over great distances (McIntyre 1988). It is likely that both residents and intruders obtain more reliable or perhaps additional infor¬ mation about each other’s fighting ability and motivation when intruders approach within 20 m of the resident pair and engage in social gatherings. Residents, in response, yodel more often (yodel ing approximately once for each such intrusion), and give more repeat phrases in each yodel. Longer yodels likely provide intruders more reliable information about the identity, health, and vigor of the signaler. However, this information comes with considerable costs to the signaler. Physiologically, longer yodels necessar¬ ily are more energetically expensive to produce. A male that produces longer yodels must be able to endure these costs. Producing longer yodels could also place the signaler at a greater risk of retaliation from the receiver. Our results have shown that longer yodels evoke quicker responses by and more tremolos (signaling greater alarm) and yodels (signaling greater threat) from con- specific receivers. These receiver-independent and receiver-dependent costs of producing longer yodels ultimately could prevent males from luffing about their motivation, and act as seective processes to maintain signal honesty (Vehrencamp 2000. Hurd and Enquist 2001) We fee both types of easts are likely more than offset by the significant costs of lighting, as aSfire„ivc pursutts end fighls (Rumntel Li S 1975, McIntyre 1988, Pnruk 2006) ettn lead to resident s evict, on or death (McIntyre 1988. pJruk 1999. Piper et al. 2008). However, more detailed analyses are needed to consider these factor, regarding fitness costs associated with pnxiucing longer yodels. We found loons yodel in the vulture posture most olten when contests escalate into social gatherings and actual chases/fights. The vulture posture may amplily/reinlorce the communication of a higher motivational state and/or it may sene as u separate signal of aggression (Rummel and Goetzinger 1978, Johnstone 1996). The posture is clearly more costly for males to assume, as they must vigorously paddle their feet to "stand" upright upon the Water surface and extend dieir wings outward while ‘pointing’ their bills toward the potential receiver (McIntyre 1988). We provide empirical support of Barklow's (1979) contention that male Common Loons signal greater aggressive motivation hy adding repeat phrases to territorial yodels. However, signaling motivation by lengthening territorial vocalizations is not the conventional way birds signal greater motivation (e.g., Morton 1977, 1982: Ripmeester et al. 2007). Many species that sing longer songs tend to communicate greater lighting ability (e.g., Lambrechts and Dhondt 1986, 1987; Appleby and Redpath 1997). In contrast, male Common Loons communicate fighting ability through the dominant acoustic frequencies ot yodels: larger males of better fighting ability produce lower-frequency yodels and smaller males of poorer fighting ability produce higher-frequency yodels (Mager et al. 2007a). Past studies have shown that males that signal lower fighting ability produce yodels with more repeat phrases (Mager et al. 2007a). We believe that males of poorer fighting ability must be more aggressively motivated (and produce longer yodels) to successfully defend their territories based on our empirical support of Barklow's (1979) hypothesis that males commu¬ nicate greater aggressive motiv ation by lengthen¬ ing yodels. This strategy may not be limited to loons, as other species similarly lengthen threat signals when more motivated to attack. Many examples are non-oscines (e.g.. Martin-Vivaldi el al. 2004), including some waterbirds (e.g.. Nelson 1984, Ewins and Weseloh 1999. Mow bray et al. 2002) living in open water environments where acoustic scatter and absorption are minimal. This raises the question regarding those selective factors associated with signal production, propa¬ gation. and reception among birds that call over Mageretal • LOONS SIGNAL HIGH AGGRESSION WITH LONGER YODELS 79 open water that would favor this means of honest communication of aggressive motivation. ACKNOWLEDGMENTS Parts of this manuscript were presented as partial Liniment of requirements of J. N. Mager's Doctor of Philosophy degree, in agreement with guidelines of IACUC approval 97-12-02 at Cornell University. Support was provided by Cornell University (Cornell Laboratory of Ornithology Walter Benning Fellowship, Department of N'eurobtology and Behavior Student Research Grant. Edna Bailey Suzman Fellowship. Kieckhefcr Adirondack pel owship. Local Sigma Xi Grunts-in-Aid-of-Research Grant. University Travel Gram), the Sigurd Olson Environmental Institute Loon Research Award, ihe Denison University Visiting Scholar Program, and the Ohio Northern l Jnivcrsity F.iculty Summer Research Grant. We thank A. A Dhondt. H. K. Reeve, and S. L. Vehrencamp for constructive recom¬ mendations in study design and manuscript preparation, kareu Grace-Martin and F. M. Vcrmcylcn lor statistical consulting, A. R. Lindsay. M. W. Meyer for assistance, the many held assistants for their hard work, and private limdown«S for access to lakes, encouragement, and support. LITERATURE CITED 'riMAXN. J. 1974. Observational study of behavior: sampling methods. Behaviour 49: 227-267. WEBV, u. M. AND S. M. Redpath. 1997. Indicators of "tale quality in the hoots of Tawny Owls (Arm 'Him Journal of Raptor Research 3 1 : 65-70. Hi!'!. | |9gg 7|,e behavioural biology of aggression, -amlifidge University Press. Cambridge. United Kingdom. •OrivUiw. W. E. 1979. The function of variations in the oculizations of the Common Loon (Gavia immer). Dissertation. Tufts University. Boston. Massachusetts. IJSA, ;^1 kv ,J. W. and S. L. Viiiupncami’, I '>98. Principles 1,1 animal communication. Simmer Associates Inc.. Sunderland, Massachusetts, USA. ' ,,f1‘ M. S. and W. A. Searcy . 1991. Acoustical WTimunication of aggressive intentions by territorial "'ale Bobolinks. Behavioral Ecology 2: 319-326. J'«tt,STEEN. T„ p. k. McGregor. J. Uni t and. J. A. foHivs. and s. B. Petersen. 1997. The signal ’“nction of overlapping singing in male Robins. Animal Behaviour 53: 249-256. •IIST. M. and O. LEIMAR. 1 9X3. Evolution of fighting Nhaviour: decision rules and assessment of relative 'length. Journal of Theoretical Biology 102: 3S7-4I0. '■wist, M. and O. Lf.imak. 19X7. Evolution of lighting behaviour: the effect of variation in resource value. Journal of Theoretical Biology 127- 187-205. >INs- P. J AND D. V. C. WKSELOH. 1999. Little Gull U-arus mbiutus). The birds of North America, dumber 428. llRr,;P L. AND M. Enquis-1 . 2001. Threat display in birds. Canadian Journal of Zoology 79: 931 -942. Jow®ON, M. 2002. Individual growth analysis using PROC MIXED. Paper 253-27 in Proceedings of the twenty- seventh annual SAS users group international confer¬ ence. SAS Institute Inc.. Cary, North Carolina, USA. Johns roNL. R. A. 1996. Multiple displays in animal commu¬ nication: 'backup signals’ and ’multiple messages.’ Philosophical Transactions of (he Royal Society of London, Series B 351:329 338 Kkoodsma. D. E. 1 989. Suggested experimental designs for song playbacks. Animal Behaviour 37 600-609. I. AMISRI.CII is. M. ANLJ A. A. DHONDT. 1986. Male quality, reproduction, and survival in the Great Tit ( Pants major). Behavioural Ecology and Sociobiology 19: 57-64. LamBRLCIITS, M. and A. A. DHONril. 1987. Differences in singing performance between male Great Tits. Ardea 75; 43-52. Lanc.i,mann, U., J. P. Tavares. T. M. Peake, and P. K. McGregor. 2000. Response of Great Tits to escalat¬ ing patterns of playback. Behaviour 137: 451—471. Li Prf.li. C. G.. m. D Hauser, and D. B. Moody. 2002. Discrete or graded variation within rhesus monkey screams? Phychophysicai experiments on classifica¬ tion. Animal Behaviour 63: 47-62. Mag HR. J. N. 1995. A comparison of the time-activity budgets of breeding mule and female Common Loons ( Gavid immer). Thesis. Miami University. Oxford, Ohio, USA. MAGER, j. N. AND c. Walcott. 2007. Structural and contextual characteristics of territorial “yodels” given by male Common Loons (Gavia immer ) in northern Wisconsin. Passenger Pigeon 69: 327-337. Magi r J. N. C. WALCOTT, and W. H. Piper. 2007a. Male Common Loons. Gavia immer. communicate body mass and condition through dominant frequencies of territorial yodels. Animal Behaviour 73: 683-690. Mager J. N. C. Walcott, and W. 11. Piher. 2007b. Nest platforms increase aggressive behavior in Common Loons. Naturwissenschaften 95: 141-147. Mager .1. N, C. Walcott, and W. H Pih-.r. 2010. Common Loons can differentiate yodels ol neighbor¬ ing and non-nbighboring conspcci tics. Journal ol Field Ornithology 81: 392-401. MARTIN. P. and P. Bateson. 1993. Measuring behaviour: an introductory guide. Second Edition. Cambridge University Press. Cambridge. United Kingdom. Martin- V i valdi. M.. J. J. Palomino, and M. Soler. ■>004. Strophe length in spontaneous songs predicts male response to playback in the Hoopoe (Upupa epops). Ethology 110: 351-362. Mayn ard Smith. J. 1982. Do animals convey information about their intentions? Journal of I heoretical Biology 97: 1-5. McIntyre. J. W. 1988. The Common Loon: spirit of northern lakes. University of Minnesota Press. Min¬ neapolis, LISA. Mowbray, T. B.. C. R. Ely. J. S. Sedjnger, and R. E. Trost. 2002. Canada Goose (Bran! a canadensis). The birds of North America. Number 682. Morion. H. S. 1977. On the occurrence and significance of motivation-structural rules in some bird and mammal sounds. American Naturalist 111: 855-869. Morton, E. S. 1982. Grading, discreteness, redundancy, and motivation-structural rules in some bird and 80 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124, No. 1, March 2012 mammal sounds. Pages 183-212 in Acoustic commu¬ nication in birds (D. E. Kroodsma, E. H. Miller, and H. Ouellet, Editors). Volume 1. Academic Press. New York. USA. Nelson. D. A. 1984. Communication of intentions in agonistic contexts by the Pigeon Guillemot. Cepphus columba. Behaviour 88: 145-189. Parker. G. A. 1974. Assessment strategy and the evolution of fighting behaviour. Journal of Theoretical Biology 47: 223-243. Paruk. J. D. 1999. Territorial take-over in Common Loons ( Gavia immer). Wilson Bulletin 111:1 16-1 17. Paruk. J, D. 2006. Testing hypotheses of social gatherings in Common Loons ( Gavia immer). Hvdrobiologia 567: 237-245. Piper. W. H . K. B. Tischlkr, and M. Kl-iCH. 2000. Territory acquisition in loons: the importance of takeover. Animal Behaviour 59: 385-394. Piper, W. H„ C. Walcott, .1. N, Macier. and F. Spilker. 2008. Fatal battles in Common Loons: a preliminary analysis. Animal Behaviour 75: 1 109-1 1 15. Piper, W. h.. C. Walcott, J N. Maghr. M. M. Perala. K. B. Tischi.er. E. B. Harrington, a. j. Turcoith, M. S< hwahenlander. and N. Banreld. 2006. Prospecting in a solitary breeder: chick production elicits territorial intrusions in Common Loons. Behav¬ ioral Ecology 17: 881-SX8. Piper, W H„ D. C. Evers. M. W. Meyer. K. B. Tischler. J. D. Kaplan, and R. C. Fleischer. 1997a. Genetic monogamy in the Common Loon (Gavia immer). Behavioral Ecology and Sociobiology 41: 25-31. Piper. W. H.. J. D. Paruk, D. C. Evers’. M. W. Meyer. K B. Tischler. M. Klich, and J. J. Hartigan. 1997b. Local movements of color-marked Common Loons. Journal of Wildlife Management 61 : 1253-1261. Ripmeester. E. A. P.. A. M. df. Vries, and H. Slabbekoorn. 2007. Do Blackbirds signal motivation to fight with their song? Ethology 113: 1021-1028. Rummel, L. and C. Goetzinger. 1975. Communication of intraspecific aggression in Common Loons Auk "2 333-346. Ri mmi i . L. and C. Goetzinger. 1978. Aggressive dispbv in the Common Loon. Auk 95: 183-196. Singer, J. D. 1998. Using SAS PROC MLXED to fit multilevel models, hierarchical models, and individu growth models. Journal of Educational and Behavior, i) Statistics 24: 323-355. SiOlander, S. andG. Argun. 1972. Reproductive behavior of the Common Loon. Wilson Bulletin 81: 296-308. van Rhijn. J. G. 1980. Communication by agonistic displays: a discussion. Behaviour 74: 284-293. V t tiRENCAMP. S. L. 2000 Handicap, index, and conven¬ tional signal elements of bird song. Pages 27'-300i ' Animal signals: signaling and signal design n aitim, communication. (Y. Esptnark. T. Amundsen, and 0. Rosenqvist, Editors). Tapir Academic Press Tmnd beim. Norway. Vehrencamp, S. L. 2001. Is song-type matching a con¬ ventional signal of aggressive intentions? Proceeding- of the Royal Society of London, Senes B 268: I637- 1642. Vogel, H. S. 1995. Individuality in. and discrimination through, the two-note wail and vodcl calls of the Common Loon (Gavia immer l. Thesis. University o' Guelph. Guelph, Ontario. Canada. Waas. J. R. 1991. Do Little Blue Penguins signal their intentions during aggressive interactions with strang¬ ers? Animal Behaviour 41: 375-382. Walcott, C„ J. N. Mager, and w. H, Piper. 2006. Changing territories, changing tunes: male loons. Gavia immer. change their vocalizations when they change territories. Animal Behaviour 71: 673-683. Walcott. C., D. Evers. M. Froehler, and A. Kr.ak.aier 1999. Individuality in ‘yodel' calls recorded from a banded population of Common Loons. Gavia immer Bioacoustics 10: 101-114. The Wilson Journal of Ornithology 124(1): 81 -86, 2012 TERRITORIAL FIDELITY TO NECTAR SOURCES BY PURPLE-THROATED CARIBS, EULAMP1S JUGULARIS VINITA GOWDA,1 24 ETHAN J. TEMELES.1 AND W. JOHN KRESS2 ABSTRACT— We present the first record of territorial site-fidelity across multiple years by Purple-throated Caribs lEulampis jugularis) on three different islands in the eastern Caribbean, St. Kitts. Dominica, and St. Vincent. Marked male Purple-throated Caribs were monitored throughout the flowering season ot their main nectar resources. Heliconia carihaea St. Kitts and Dominica) and H. hihai (St. Vincent), both native perennial herbs. Individual males were observed defending the tat Heliconia patches for 3 years (St Vincent). 4 years (St Kilts). and 5 years (Dominica), and remained near these patcho e.en when they were not in flower. The territorial behavior and resource dependence of Purple-throated Caribs on native heliconias likely have a key role in the coevolution of this specialized plant-pollinator interaction. Received 17 Man h 201 1. Accepted 2 August 2011. Territoriality is a costly and often complex behavior involving exclusive possession and/or defense of an area against conspecific and hetero- specific animals (Carpenter 1958; Brown 1964. 'W; Pyke 1979). Territorial defense of food resources is common by nectar- feeding birds (Gill and Wall J975,Cruden and Hermann-Parker 1977, tf 100 ramets) in a given year (Berry and Kress 1991). Each ramet bears one multi¬ tiered inflorescence which consists of several sequentially opening flowers within a bract and several sequentially opening bracts within an inflorescence. Each flower within a bract lasts only a single day and flowers remain open from dawn to dusk (Temeles et al. 2005, Gowda and Kress 2008). Puqile-throated Caribs were banded at each site (Table 1) during March and April with unique color-coded darvic bands (www.avinet.com) and observed for 24 hrs (8 hrs/day X 3 days) per month in March. April, May, June. July, September. November, December, and January . These months represent both the peak (Mar to Jul) and the non- peak (Aug to Feb) flowering seasons of the two native heliconias (Gowda 2009) and also cover tie breeding period of Purple-throated Caribs (Mar i>> Jul) (Wolf 1975, Temeles and Kress 2010). All observations of territoriality by Purple- throated Caribs on St. Kills and Dominica were made at marked clumps of H. caribaea plants over multiple years. H. caribaea is very rare on St, Vincent (only 2 clumps found in the forest); thus, observations of territoriality by Purple-throated Caribs wen conducted on marked clumps of H. bihai. We delineated the boundaries of territories by noting the point at which territorial males evicted intruders, which included both conspecific males and females, and heterospecifics (Temeles et al 2005). We concluded that males were display¬ ing ‘territorial fidelity' if the same marked territorial bird was observed and recaptured at the same marked feeding plants for >2 yrs after first capture. We also recorded the number of flowers within the clump on each observation day to assess use of Heliconia patches by males in relation to flowering. RESULTS Territoriality at Heliconia caribaea.— We ob¬ served few visits (4-5 visits/day) by male Purple- throated Caribs to clumps of H. caribaea in the beginning of March when budding inflorescences were visible but no open flowers were available Male Purple-throated Caribs on all islands, but especially on Dominica, increased their visitation frequency as the flowering season progressed (Fig. 2) and spent more lime on the territory defending against incoming visitors to single flowering inflorescences or often even at inflo¬ rescences without flowers. Male Purple-throated Caribs rarely left the vicinity for more than a fe'v hours a day or rarely for an entire day dunng the beginning of the flowering season of H. caribaea indhdditfs momWdn' !*! °' ' PurplcM,m>aled Caribs from three eastern Caribbean Islands. Number - 1 _ PU lsljnd "■ individuals displaying territoriality = T. and individuals displaying site fidelity = Island Territorial months on non -Hdiconia plan's Gowda el al. • TERRITORIAL FIDELITY OF THE PURPLE-THROATED CARIB 83 y c. X 0 X 3 •C 0 T3 C 0 35 30- 25- 20- 15- 10- * 3 •a y a. 0 35 (A) Dominica 1 30 1 Q. X X Q (B) St. Kitts X \ X 300 250 200 150 100 50 0 -300 o 03 3 =3 -250 C 3 a- - 200 o ~t c —5 - 150 o £ - 100 o C/3 -a -50 o -n a. 03 - 0 D Chases per hour Fin in . . u. PmT>lc-throated Caribs at H. cariboeo and H. bihoi clumps • fcX, vistts and observed cha*s b m . Purple .hr=m ' vjs,,s (open) and chases ■:Z “T a'Tr h * error. The dotted line represent rhe mean Ln 3S month across a single year, rtritowfa ™ibaea on St. Kitts and Dominica, and H. MW on " vincent*, H two male Purple-throated Caribs in Domin- bv mid-April, the peak flowering season of H. ica were not observed on their territory on two caribaea , and allowed only females to feed within 'fparate davs Male Purple-throated Caribs territories, which often consummated in a mating st)0Wed aggressive territorial defense (Fig. 2A) event. Conspecific females, males, and heterospecifics 84 THE WILSON JOURNAL OF ORNITHOLOGY . Vo/. 124, No. 1, March 2012 (Green-throated Carib [Eulampis holosericeus], Antillean Crested Hummingbird [Orthorhynchus cristatus], Lesser Antillean Bullfinch [LaxigilUt noctis], Bananaquit | Coereba fluveola]) were chased from the territory by the resident male Purple-throated Carib. Displacement of a resident male Purple-throated Carib was observed only once during the study (Dominica. Carholme site). A new male Purple-throated Carib displaced the resident focal male in the second year of observation after a rapid aerial chasing and scrub- level scuffle. Territorial defense at H. caribaea by male Purple-throated Caribs lasted from March to July on Dominica (Table 1). Territorial defense had higher temporal variation on St. Kitts and St. Vincent (Fig. 2B, C; Table 1). For example, on St. Kitts, an all-day territorial presence at H. caribaea clump by Purp I e- throated Caribs was observed only from March to May, whereas territorial defense by Purple-throated Caribs on St. Vincent was observed only in April and May and was further restricted within a day, between 0900 to 1500 hrs (Fig. 2C). Territorial defense by females was rare and only observed on St. Vincent where females chased conspccific females (4 times) and nonspecific males (2 times). However, females chasing males could not be distinguished from their mating repertoire. Territorial Fidelity at Heliconia Patches.— Ail marked males on the three islands were faithful to their feeding territories across multiple years (Table 1). Individual males remained faithful to their Heliconia patches for 3 years on St. Vincent, 4 years on St. Kitts, and 5 years on Dominica. Females did not show territorial behavior during the breeding season but the same female was recaptured in 2 years in the same feeding clump of H. bihai and H. caribaea on Dominica. Similarly, a banded, non-territorial female Purple-throated Carib was repeatedly observed at the same patch of H. bihai on St. Lucia over 2 years (EJT, unpubl. data) and in St. Vincent for 2 years. Thus, females may not defend territories, but thev apparently exhibit site fidelity to Heliconia patches when traplining. Purple throated Carib Territoriality and Site P Hi"' ^ Absence of Heliconias.- Male Purple- throated Caribs exhibited territorial behav- ZeTof ,h ofkothcr P,mt *peci« during tunes of the year when H. caribaea was not (lowering. We observed the birds defending Citrus P" US SP" Ge**eria cymosa, Heliconia psittacorum, H. wagneriana , Inga ingoides, Lobe¬ lia cirsiifolia, Marcgravia umbellate. Musa sp.. and Wercklea tulipiflora. Seven of nine resident males, despite the complete absence of floweraof H. caribaea from August to March, were observed in the general vicinity ol their H. caribaea territory in September, November. December. and January on St. Kitts (2 of 3 birds) and Dominica (5 of 6 birds). One or more of the non -H. caribaea species was present within 200-1.000 m of Heliconia patches for all of these males. Male Purple-throated Caribs shitted their territorial defense to these plant species, although one of the marked birds was observed in a citrus farm >1 km distant. DISCUSSION Territoriality, foraging, and mate-selection in hummingbirds are behaviors strongly known to be influenced by local energy sources (Carpenter 1958. Gass et al. 1976, DeJBenedictis et al. I97S. Gass 1978, Kodric-Brown and Brown 1978; Montgomerie el al. 1984. Gass and Sutherland 1985. Temeles and Kress 2010). The abundant heliconias in the eastern Caribbean Islands repre¬ sent critical sources of nectar for hummingbirds due to the: ( I ) clonal habit of the plant that assure" high density within a small radius (2 to 3J individuals/m*), (2) nectar rich flowers within the same inflorescence (up to 10 to 12/inflorescence), and (3) long flowering season lasting several months (Apr to Jul) that assures a stead}, dependable food source (Temeles et al. 2005. Gowda 2009, Temeles and Kress 2010). The presence of territoriality by Purple-throated Caribs throughout the flowering season of the heliconias and their associated multi-year site-fidelity con¬ firms the two heliconias (H. caribaea in St. Kilt' and Dominica, and H. bihai on St. Vincent) at* critical nectar sources for Purple-throated Carib' The presence of the same individuals at specific Heliconia clumps within a flowering season arid across years suggests the Purple-throated Caribs on these islands have a long-term imprint of high- quality loraging sites that are not abandoned eiihct within or between years. Hummingbirds use both fine (Miller et al. 1985, Sutherland and Gass 1995 and coarse-grained memory' (Armstrong et at 1987) to locate optimal food resources. Purple- throated Caribs likely use coarse-grained spatial memories to re-defend the same clump of Li caribaea year after year (Baida and Kamil 1989). Our observations of Purple-throated Caribs in the genera! vicinity of their territorial sites during Gowda et aL • TERRITORIAL FIDELITY OF THE PURPLE-THROATED CARIB 85 the non -H. caribaea flowering season suggest they may be aiding their spatial memory by main¬ lining site-fidelity even when heliconias arc not in Hotter. On a broader level, many migratory North American hummingbird species have been a-capiured at the same general locations between successive years (Bassett and Cubie 2009), and vitne lekking hummingbird species return to the sime leks between years (Stiles and Woll 1979). Long-term spatial memory may be a general characteristic of hummingbirds, although it may have evolved through a variety of different con¬ texts (feeding, migration, and mating). Our observation of strong fidelity to Helicon ia patches by Purple-throated Caribs has implica¬ tions for coevolution between hummingbirds and Hcliconia. Bill morphology, body size, and energetics of male and female Purple-throated < aribs arc closely related to one or the other species of Heliconia on the islands of St. Lucia and Dominica (H. caribaea with males: H. bihai with females; Temeles et al. 2000, Tcmeles and Kress 2003). Close correspondence between hummingbirds and flowers suggests these islands ^ 'hotspots’ (Thompson 2005) of reciprocal evolution. Thompson (2005) noted that gene flow among populations can dilute and weaken recip- 10031 adaptation. However, year-to-year fidelity "• territorial male Purple-throated Caribs to the '•'me patches of heliconias combined with territo- "Ol exclusion of intruding conspecifics and hetero- •c-ecifics would reduce gene flow in both plants and hummingbirds, resulting in strong coevolution of ^Tic-throated Caribs with //. caribaea through toiprocal selection for flower numbers and male tod fighting ability (Temeles and Kress Similar fidelity by traplining females may ‘^-'contribute to strong ecological interactions and 1 duuonary interactions with II bihai through flower size and shape and bill morphology 'Temeles et al. 2009). Benkman et al. (2003) sported a case of coevolutionary interactions fvttteen a non-migratory population ot Red Cr,Abil|s [Loxia cunirosira) and lodgcpole pine lp'n“s contorta spp. latifolia) where reciprocal ■election has largely shaped their bill morphology ^ ^rd defenses, respectively. Our observations indicate that male Puiple- 'hroated Caribs continue to associate with patches caribaea even when these plants are not in This raises the questions of whether males *** defending and exhibiting fidelity to the Hcliconia species, or to the sites in which these plant species occur. We suggest it may be both and have no doubt that heliconias arc the resource magnet that attracts hummingbirds to these sites. Our contention is supported by the observation that with the exception of introduced plants (e.g., bananas, citrus, and some ornamentals): we have found no native plant species on these islands that are as rewarding as heliconias in terms ol overall nectar production (flowers of these heliconias produce 60 to 300 pi per day: Gowda and Kress 2008). Thus, fidelity to a long-lived, highly- rewarding resource, even when it is not in flower, may be a viable territorial strategy, especially if alternative low-quality resources are locally avail¬ able. Once a territory is abandoned, it may be energetically more costly to re-establish it than to maintain it during times of low flower availability. acknowledgments Wo thank the Minisiry ot Agriculture and the Environ¬ ment; Forestry. Wildlife, and Patks Division; Common¬ wealth of Dominica. Government of St. Christopher and Nevis, and Forestry Department. Ministry of Agriculture and Fisheries, St. Vincent and the Grenadines tor facilitating this study. We also thank Arlington James. Enc 1 1\ polite. Ida Lopez. Anne Jno Baptiste, and Gregory Pereira for logistic help and discussion. Wc thank C. E. Braun and two anonymous reviewers lor helpful comments and assistance on this manuscript. This work was supported by a Weiniraub Graduate Fellowship from the Department of Biological Sciences. The George Washington Univers.ty to VG and also by the Cosmos Club. Sigma-Xi. and the Smithsonian Institution, Field work for ibis study was supported by grants from National Science Foundation (DEB-06142 IS) to E. J. Temeles and W, J. Kress. LITERATURE CITED Armstrong. D. 1992. Conelation between nectar supply and aggression in territorial honeyealers: causation or coincidence? Behavioral Ecology and Sociobiology 30:95-1 02. Armstrong. D. A.. C. L. Gass, and G. D. Sutherland 1987. Should foragers remember where they ve been. Explorations of a simulation model based on the behaviour and energetics of territorial hummingbirds. Puses 563-586 W Foraging behaviour (A. C. Kamil. J. R. Krebs, and H. R. Pulliam. Editors). Plenum Press. New York, USA. Baloa. R. P. and A. C. Kamil. 1989. A comparative study of cache recovery by three corvid species. Animal Behavior 38:486-495. Bassett. F. and D. Cubie. 2009. Wintering hummingbirds in Alabama and Florida; species diversity, sex and age ratios, and site fideliiy. JoumaJ of Field Ornithology 80:154-162. Benkman. C. W„ T. L. Parchman. A. Fa vis. and A. M. Siepielski. 2003. Reciprocal selection causes a 86 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. 1. March 2012 coevolutionary arms race between crossbills and Iodgepole pine. American Naturalist 162: 182-194. Berry, F. and W. J. Kress. 1991. Heliconia: an identification guide. Smithsonian Institution Press. Washington, D.C., USA. Brown, J. L. 1964. The evolution of diversity in avian territorial systems. Wilson Bulletin 76:160-169. Brown, J. L. 1969. Territorial behavior and population regulation in birds: a review and re-evaluation. Wilson Bulletin 81:293-329. Carpenter. C. R. 1958. Territoriality: a review of concepts and problems. Pages 224-250 in Behavior and evolution (A. Roe and C. G. Simpson. Editors). Yale University Press. New Haven, Connecticut, USA. Cotton, P. A. 1998. Temporal partitioning of a floral resource by territorial hummingbirds. 653. UMN i^u:o*4/ Cruden, R. w and S. M. Hermank-Parker. 1977. Defense of feeding sites by orioles and Hepatic Tanagers in Mexico. Auk 94:594-596. DeBenedictis. P. A.. F. b. Gill, F. R. Hainsworth. G. H. Pyke, and L, L. Wolf. 1978. Optimal meal si/e in hummingbirds, American Naturalist 112:301-316, Dobkin. D. S. 1984. Flowering patterns of long-lived Heliconia inflorescences: implications for visiting and resident ncctarivorcs. Oecologia 64:245-254. Evans, M. R. 1996. Nectar and flower production of Lobelia lelekit inflorescences, and their influence on territorial behaviour of the Scarlet-tufted Malachite Sunbird ( Nvchmnia juhnsmni). Biological Journal of the Liuncan Society 57:89-105. Frost. S. K. and P. G. H. Frost 1980. Territoriality and changes in resource use by sunbirds at Leonoiis leonums (Labiatae). Oecologia 45:109-1 16. Gass, C L. 1978. Experimental studies of foraging in complex laboratory environments. American Zoologist 18:729-738. Gass C. L. and G. D. Sutherland. 1985. Specialization by territorial hummingbirds on experimentally cli¬ nched patches of flowers: energetic profitability and learning. Canadian Journal of Zoology 63:2125-2133. Gass, C. L.. G. Angehr, and J. Centa. 1976. Regulation of food supply by feeding territoriality in the Rufous Hummingbird. Canadian Journal of Zoology 54:2046- Gill. F. B. and L. L, Wolf. 1975. Economics of feeding territoriality in the Golden- vs inged Sunbird. Ecology 56:333—345. Gowda, V. 2009. Pollination biology and inter-island geographical variation in the mutualistic Heliconia (He I icon i aceae)-h u m m i ngbird (TrochiUdae) interac¬ tion of the eastern Caribbean Islands. Dissertation George Washington University. Washington, DC Gowda V. and J. W. Kress. 2008. Nectar characters of two eastern Caribbean heliconias tHelico, on iaXTa ICI' coniil Socic,y Intcrn;ltio^‘ Bu Kodric-Brown, A. and j. H. Brown. 1978 Influence economics, interspecific competition and Z dimorphism on territoriality of migrant Rufous Hummingbirds. Ecology 59:285 296. Kress. W. J. 1983. Scif-incompatibijjty in Central American Heliconia. Evolution 37:735-744. Limiart, Y. B. 1973. Ecological and behavioral detaroi- nants of pollen dispersal in hummingbird-pollinated Heliconia. American Naturalist 107:511-523. Miller, R. S., S. Tamm, G. D. Sutherland, and CL Gass. 1985. Cues for orientation in huntminghml foraging: Color and position. Canadian Journal nf Zoology 63:18-21. Montgomerie. R. D„ M. J. Eadif., a\d L. W. Hardfk 19X4. What do foraging hummingbirds maximite? Oecologia 63:357-363. PYKE. G. H 1979. The economics of territory size and time budget in the Golden-winged Sunbird. American Naturalist 114:131-145. Stiles, F. G. 1975. Ecology, flowering phenology, and hummingbird pollination of some Costa Rican Wei/- count species. Ecology 56:285-301. Stills. F. G. 1978. Possible specialization for humming- bird-hunting in the Tiny Hawk. Auk 95:550-553. S t it ns. F. (i. AND L. L. Woli . 1979. Ecology and evolution °l •‘-'k mating behavior in the Long-tailed Hermit Hummingbird. Ornithological Monographs 27, Sutherland, G. D. and C. L. Gass. 1995. Learning and remembering of spatial patterns by hummingbirds Animal Behaviour 50: 1273-1286, I'EMBLES. E. J. and W. J. Kress. 2003. Adaptation in a plant-hummingbird association. Science 300:630-633 Temi-.lks. E. .1. and W. J. Kress. 2010. Male choice and male competition by a tropical hummingbird at a floral resource, Proceedings of the Royal Society of London. Series B 277:1607-1613. Jr mi i is. E. J.. R. s. Goldman, and A. U. Kudla. 2005 Foraging and territoiy economics of sexually dimor phic Purple-throated Caribs (Eulampis jugl&wO on three heliconia morphs. Auk 122:187-204. Temei.es, E. J., I. L. Pan, J. L. Brennan, and I- N. Horwitt, 2000. Evidence of ecological causation ol sexual dimorphism in a hummingbird. Science 289:441-443. Temeles, E. J„ K. C. Shaw. a. U. Kudla, and S. E Sander. 2006. Traplining by Purple-throated Carib Hummingbirds: behavioral response*; to competition and nectar availability. Behavioral Ecology and Sociobiology 6 1 : 1 63- 1 72. Temeles, E. J„ C. R. Koulouris. S. E. Sander, .and W. J Kress. 2009 Effect of flower shape and size on foraging performance and trade-offs in a tropical hummingbird. Ecology 90:1 147-1 161. Thompson. J. N. 2005. The geographic mosaic of coevolution. University of Chicago Press. Chicago. Illinois. USA. Wolf. L. L. 1975. “Prostitution" behavior in a tropica) hummingbird. Condor 77:140-144. Wolf, L. L. and F. R. Hainsworth. 1971. Time and energy budgets of territorial hummingbirds. Ecology 52:980-988. The Wilson Journal of Ornithology 1 24( 1 ):87— 95, 2012 BREEDING AND FORAGING VARIATION OF THE PLUSH-CRESTED JAY ( CYANOCORAX CHRYSOPS) IN THE BRAZILIAN ATLANTIC FOREST ANGELICA MARIA KAZUE UEJ1MA,1' ANDREA LARISSA BOESING.24 AND LUIZ DOS ANJOS2 ABSTRACT.— We monitored six flocks and five active nests of the Plush-crested Jay (Cycmocorcix chnsops) at three in the Atlantic Forest in southern Brazil. The sites had different vegetation compositions and spanned different levels i thropogemc disturbance. Home ranee size in full/wimci was 20-30 ha and the breeding territory size in spring/summer 5-10 lu in size. Territories were smaller across sites with higher anthropogenic food supplementation. Hock sizes were 5-11 individuals during spring/summer and 8-15 individuals during lall/winter. The Plush-crested Jay is a cooperative breeder, nesh were 4-7 m above ground level, the incubation period WB# 18*20 days, brood si/e (\ ± SD* was 3.4 ± 0.80 vesper nest, and nestlings fledged 23 T 1 2b days after hatching. This species occupies all forest strata but tends to use the underaory -and middle levels most (G = 178.2; P < 0.01). Invertebrates were the most frequently consumed item in all -rSfi. but percent consumption varied among sites (G = 105.06; P < 0.01). We observed 110 >ood caching events throughout the year, primarily seeds at Araucaria august (folia, maize, and coconuts. Caches were on the ground in - 40). m epiphytes (n = 47). and on branches in - 23). Levels of anthropogenic food supplementation resulted in variation in territory and home-range size, nestling survival rates, strata occupation, and diet composition of the Plusli-crcsted Jay. Romed 6 tehruary 2011, Accepted 29 August 2 Oil. Ihe family Corvidae is represented in South America by the genera Cyitnoconix and Cyanolyca i Madge and Bum 1994). These genera occur across dte neotropical region and include 17 species of ^‘inocorax and nine species of Cyanolyca (Gill “d Donsker 2011). The Plush-crested Jay ( Cya - '"wruv chrysops) is widely distributed throughout l'0Ul^ America from Amazonia to northeastern Argentina occupying a variety of habitats that l llude >everal types of forests: mixed rainforest, -mi-deciduous foresi, Cerrado (one of the largest :al savannas of the world) (Silva and Bates - "'-I. scrub, and also semi-urban areas. It is not "u'ly found in the interior of primary forest, but to be more frequent at the borders and in -^life's (Goodwin 1976. Anjos 2009). It is poorly l^eations. — Field observations were made 6 days per month in each area from 87 88 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 FICi. I Study areas in southern Brazil (Parana State) where Plush-crested Jay flocks were monitored: VVP (Vila Velh. State Park), KEP (Klabin Ecological Park), and RTF (Ribeirao do Tigre Farm). October 1995 to December 1996. We used 8 X 35 binoculars for observations that began at sunrise and lasted all day. Field observations were conducted during 1.800 hrs. The daily procedure was to walk along previously estab¬ lished transects to find one flock, anti lo follow this Hock throughout the day. Particular meth¬ ods were used to document breeding and foraging behaviors. The birds were not banded but we were able to distinguish different flocks (at least 2) at each site as American jays maintain an area of dominance around active nests (Anjos et al. 2009). We followed Gill and Donsker (201 1 ) for nomenclature of birds and the Missouri Botanical Garden (2010) for nomenclature of vegetation. Flock Size. Home Range, and Territory. — Wc recorded the number of individuals in the Hocks :i; the different sites and periods. We separated the year into two periods: the first when aggressive territorial behavior was observed between differ¬ ent flocks (the territorial period), and the second when aggressive behavior was not observed (the home-range period). We plotted each observation of Plush -crested Jay flocks on each transect onto a map (Luginbuhl et al. 2001) for both territorial and home-range estimates. We used these obser¬ vations to calculate the size of the overall territorv VJA.BI,5 C|iaracienstics of areas where Plush-crested Jays were monitored in the Atlantic Forest, southern Bra/- V.la Velha Slate Park I VVP), Klabin Ecological Park (KEP). and Ribeirao do Tigre Rum (RTF). vvp KEP Vegetation type* No. tree species** Dominant tree species Climate* Average temperature* Average rainfall* Coordinates Total area Sample area Forest cover Grassland Human built Agriculture RTF Mixed rainforest (MF) 350 species: 13% endemic Araucaria cingustifolia. Hex paraguariensis, Camponesia xanthocarpa, Ocolea porosa, O. odorfera Cft> >8 C 19.5 =c 1,550 mm 1.700 mm 25 15' S, 50 05' W 24° 17' S, 50 ’35' W 3,790 ha 11,196 ha 150 ha 23% 60% 17% 140 ha 60% 30% 10% Seasonal semi-deciduous forest (Sf 220 species; 10% endemic Aspidosperma polyneuram. Cedrt fissilis, Balfourodendron riedelianum Cfa 21 C 1,600 mm 23° 27' S. 51° 15' W 1.285 ha 150 ha 20% 20% * Mcndonya and Danni- warin: Cfa = subtropical — — - - - - - 60% humid w7th warm summer-.111* Ro,la 1 ^kf = subtropical humid with warm summer t Uejimaet al. • BREEDING AND FORAGING OF THE PLUSH-CRESTED JAY 89 and home range of the jays using the minimum convex polygon method (MCP) (Mohr 1947, Odum and Kuenzler 1955). This consists of joining the outermost observation points for each flock with a straight line. The largest polygon obtained was taken as the size of the territory of a flock. This procedure was chosen due to its simplicity and wide use in ornithology (e.g.. Jullien and Thiollay 1998. Wiktander el al. 2001. Ribeiro et al, 2002. Duca et al. 2006). Reproduction.— Reproductive activities were observed dunng -720 hrs of field observations (240 hrs/area). We recorded the number of eggs hid and hatching data, and monitored the development of nestlings. Nestlings were mea¬ sured the second week post-hatching. A nest was considered depredated when the entire brood disappeared or when there was no evidence inside or underneath the nest of any other kind of loss. Activities related to incubation, provisioning of the brooding bird or nestlings, and cleaning of the nesf 'verc also recorded. Two fledglings were color banded. The nests were measured (height, diameter, brood chamber diameter and diameter ' ivv'rs used in its construction) alter nesting activities had ceased, collected, and deposited at the Capuo da Inibuia Museum ot Natural History tC. uritiba. Brazil) and the Klabin Ecological Park Museu,n (Telemaeo Borba, Brazil). paging.- We made field observations of 'paging behavior during 1.080 hrs (360 hrs/area). '• used focal animal sampling following Altmann with 5-min breaks between observ ation of ?erem individuals or flocks (scan-sampling). procedure should have increased data inde- penT^nce. The food item, capture substrate, and "“lum were recorded for each foraging event of 1,1 individual jay. We considered three levels ol strata, besides the ground: undergrowth (up Jj1 - m above the ground), middle level (from 2 to ll: and suheanopy (from 7 m above the ground 1 'ii below the canopy). The substrates were: c! Jun{T branches, leaves or epiphytes, and air lor foraging. . Statistical Analyses.— The R X C test for dependence (G-test) was used to evaluate the Mgnificance (P = 0.01) of proportions of events dated to foraging behaviors among the three sites '.Fowler and Cohen 1986). RESULTS Rock Sizes, Home Range, and Territory — lerritorial behavior was first observed in early October (spring). The size (x ± SD) of flocks during this period was smaller at VVP (6 ± 0.8 individuals) and at RTF (5 ± 0.2 individuals) than at KF.P (II ± 0.6 individuals). Territories were smaller at VVP (5 ± 0.3 ha) and KEP (5 ± 0.7 ha) than at RTF (10 ± 0.4 ha: Fig. 2). Territories at KF.P and RET did not overlap and were separated by 350 and 500 m. respectively. However, over¬ lapping (~ 20 %) of territories was detected at VVP. The area of greatest anthropogenic food supplementation had the greatest overlap at VVP, and aggressive territorial behavior was frequently observed at this site. Aggressive territorial behavior between individuals started with “bob¬ bing", (described by Hardy 1961). followed by alarm calls. Direct contact between individuals was recorded when the dominant individual touched the back of an opponent after sweeping flights. At times, the opponent persisted and the two individuals attacked each other with their feet while the remainder of the individuals in the (lock stayed close, emitting alarm calls. Territorial behavior was relaxed throughout the end of summer (Mar). Flock sizes were largest in the home-range period (Apr-Sep, mostly fall/wintei ) at KEP ( 15 ± 1.92 individuals), followed by RTF (11 ± 1.48 individuals), and VVP (8 ± 1.63 individuals). Home ranges were smaller at VVP (20 ± 1.34 ha) than at KEP (30 ± 0.89 ha) and RTF' (30 ± 1 .04 ha; Fig. 2). Overlapping of home ranges (where aggressive behavior was not recorded among individuals ol difterent flocks) was —25% in all areas. Reproduction.— Five active nests were found, three at VVP and two at KEP. The first active nest was found in mid-October at VVP. Nests were placed closer to the animal breeding center at KEP and to the restaurant at VVP. Only one nest was successful at VVP. The other two nests found in this area were depredated several times (n - 4) after hatching. Nests were reconstructed (each 1 8 to 21 days after depredation) into February, but none was successful. Two new nests were 50 m from the previous nests and the other two were in trees that were closer, but no new nests were found in the same tree in which a nest was pre¬ viously depredated. Nests measurements (mm) in the Atlantic Forest, southern Brazil: Klabin Ecological Park (KEP) and Vila Velha State Park (VVP). Nesi U Nest height Nest diameter 1 KEP 2 VVP 3 VVP 4 KEP 5 KEP Mean ± SD Brood chamber height Brood chamber diaroMW 152 180 190 125 108 152 ± 31.27 287 282 275 290 343 295.4 ± 24.33 99 120 108 90 80 99.4 ± 13.88 180 170 162 150 148 162 ± 12.07 Uejima et al • BREEDING AND FORAGING OF THE PLUSH-CRESTED JAY 91 FIG. 3. Eggs of Plush-crested Jays in Klabin Ecological Park (KEP) in southern Brazil. Photograph by A. F. R. Gatto. and offspring care. Food was brought to the brooding bird by helpers in the flock after constant calling for food. The helpers either directly delivered food to the nestlings or to the lJl|ll in the nest; this individual received the food Jl|d redistributed it to the nestlings. The average dumber of visits to provision nestlings varied ’froughout their development: the average was t;'c visits/hr in the lirsl week; seven visits/hr in L second week, and nine visits/hr in the third l-' The average visitation rate was the same -Prdless of brood size. Helpers were recorded "uing fecal sacs from the nest and providing (ic!ense against predators in addition to helping Pjov'ision the nestlings. Up to six helpers were :,early observed defending a nest against tufted capuchins ( Cebus cippelci), which were kept far front the nest. Nestlings left the nest ~23 ± 1.26 days after hatching and, for the next 20 days, they solely depended on food provided by the parents and helpers. Fledglings began feeding by themselves after 25 days and were completely independent at 90 days. The two color banded fledglings marked in 2005 were observed helping in 2006. Foraging.— We observed 3.508 foraging events (1.209 at VVP; 1,700 at KliP; and 599 at RTF; Table 4). The capture technique used by Plush- crested Jays was ‘gleaning' (described by Robin¬ son and Holmes 1982). The diet (recorded in all 3.508 events) included invertebrates (spiders, millipedes, beetles, caterpillars, grasshoppers, I KEP 2KEP 3 KEP 3 KEP 5VVp 6 Wp 7 KEP 8 KEP ’KEP Mean ± SD Tarsus 38.8 36.4 36.5 39 38 36 39.1 36.8 40.5 37.9 ± L46 Cut men 20.2 18.2 19.4 20.6 15.2 19 17.6 17.5 18.2 18.4 ± 1.53 Noslril/Bitl Wing 12.2 92 11 80 12.7 89.3 12.1 89 12 110 14.1 110 10.7 82.4 11.3 82.4 11.1 82.5 11.9 ± 0.99 90.8 ± 10.90 Mass 140 115 120 130 100 100 115 ± 12.47 92 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 4p wwof £?Tency (,PerCenIage) °f f°°dS COnSUmed by adu,t Plush-crested Jays in southern Brazil: Vila cilia State Park (VYP). Klahin Ecological Park (KEP), and Ribeirao do Tigre Farm (RTF). Food item VVP KEP RTF Total Insects/ invertebrates External resources* Fruit Vertebrates Totals 964 (80) 134 (11) 31 (3) 80 (7) 1.209 (100) 1.457 (86) 103 (6) 89 (5) 51 (3) 1.700 (100) 559 (93) 21 (4) 1 1 (2) 8 (1) 599 (100) 2,980 (84.9) 258 (7.4) 131 (3.7) 139 (4) 3.508 (100) * Garbage, maize seeds. and mol I asks), fruit, many kinds of foodstuffs given or discarded by humans (garbage, bread, meal, cookies, grain plants, and maize), and vertebrates (frogs, eggs, and nestlings, and small reptiles). Invertebrates were the most commonly used tood resource at all ihree areas, but the percentages of each item varied (G = 105.06; df = 6; P < 0.01; Table 4). Food resulting from human activities was used more at VVP and KEP than at RTF where the most commonly consumed items were insects and invertebrates. Plush-crested Jays fed on fruit from 24 plant species, but only the fruits of Syagrus roman- zofficma were commonly eaten in all three are,;' ( fable 5). A greater variety of fruit was observed to be consumed at KEP ( 13 species). while 10 and five Iruit species were recorded at VVP and RTF. respectively. The fruits most frequently recorded were: Eriobotrya japonica , Diospyros kaki. Phil¬ odendron spp., Casearia sylvestris. Rapanea ferruginea, Syagrus romanzoffianum, and Fine enormis. TAELE 5. Plant species used by Plush-crested Jays and relative frequencies of fruits consumed in southern Brazil- “ ,VVPX ** (KEP), and Ribeirao do Tigre Farm (RTF). Plant species VVP Urera haccifera Psidium cattleianum Paullinia carpopoda Lepismium cruciforme Ficus enormis F. insipida F. monckii Melia azedaracli Syagrus romanzoffiana Casearia sylvestris Rapanea ferruginea R. umbel lata Philodendron spp. Cestrum calycinum Miconia spp. Sty rax leprosus Citronella gongonha Morus nigra Trema micrantha Diospyrus kaki Citrus spp. Eriobotrya japonica Peschiera australis Celt is iguanaea Totals KEP RTF 3 4 2 1 3 0 2 1 4 7 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 12 15 6 14 4 1 I 1 6 1 9 0 13 0 0 85 0 0 0 0 0 I 0 I 7 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 12 The foraging forest stratum and capture sub¬ strate were recorded at 2,980 foraging events (964 VVP, 1,457 KEP, 559 RTF; Table 6). 'Hie pro¬ portion ol observations related to foraging stratum differed among the areas studied (G = 178.2; df =* 6; P < 0.01). The understory and the middle level were the most frequently used strata in all areas, but the ground and the subcanopy were proportionally used more at VVP and RTF respectively. The proportions of use of the capture substrate' also differed among the three areas (G - 80.84. df = 8; P < 0.01). Branches and leaves were generally the most frequently used substrates in the understory and the middle level in all art4> (C = 24.18; df = 6; P < 0.01): epiphytes were less used, particularly in the subcanopy, and the air was the least used. Food caching behav ior was also observed and recorded (n = 110). Each individual had its own cache up to 100 m from where the food was captured. Each cache contained only one unit ot a given item (except for maize, for which each cache could contain up to 12 seeds). Leaves or small sticks were laid over the food cache. We did not observe individuals defending their caches. Cach¬ es locations were in the ground ( n — 40), in epiphytes (// = 47). and on branches (n = 23). The caches mainly contained seeds of .4. angustifolia, bones, maize, and seeds of 5. romanzoffianum). Uejima el al. • BREEDING AND FORAGING OF THE PLUSH-CRESTED JAY 93 TABLE 6. Number of feeding events by strata and aging substrate of Plush-crested Jays at each of the study lies in southern Brazil: Vila Velha State Park (VVP), Klabin Ecological Park (KEP), and Ribeirao do Tigre Farm l RTF I. Sc* VVP KEP RTF Totals Indemon Branches 164 204 56 424 Leaves 132 187 70 389 Epiphytes 81 106 30 217 .Air 8 17 13 38 Subtotals 385 514 169 1.068 Middle level Branches 122 168 89 379 Leaves 138 224 90 452 Epiphytes 82 137 39 258 Air 6 17 8 31 Subtotals 348 546 226 1.120 Subcanopy Branches 11 118 42 171 Leaves 17 93 56 166 Epiphytes 17 32 18 67 .Air 0 12 6 18 Subtotals 45 255 122 422 Ground 186 142 42 370 Totals 964 1.457 559 2.980 Ihe number of foods cached did not differ between seasons (G = 2.06; df = 3; P > 0.01 ). DISCUSSION ' l;lrger size of the territory at RTF than a 'pand KEP was probably related to the amount 01 SuPplementary food at the two latter sites. L:ir^ territory sizes at RTF may compensate for 'u' *vith lower food availability. Marzluft and VJlherlm (2006) observed that crows and ravens 1 1 ' spp.) had smaller home ranges and higher ambers of fledglings near settlements. However, 1 tound larger home ranges at KEP than at VVP. ;ilhougb nock Size was similar. This may be due l| higher predation rates at VVP (lower fledgling Auction) than at KEP. allowing a smaller home r^e- Territory size expanded at all sites during !!,c summer (Fig. 2). probably due to addition of iledglings to flocks or possible variation in food abundance. of the available data published on •oiiocorux species indicate communal bleeding "[Un active nest in each group and helpers in IL‘ nest during the nesting period and while the .’"iing are dependent (Anjos el al. 2009). We documented this system in our study of the Plush- crested .lay. Alternation among individuals was observed during the incubation period, unlike for most Cyanocorax jays, where only one individual incubates (e.g.. Inca Jay |C. yncas |: Alvarez 1975). This behavior appears similar to that of the Curl-crested Jay (C. crisiatellus), where short substitutions were observed (Amaral and Macedo 2003). Two individuals were observed incubating for Azure Jays (C caendeus). at times with relief (Boesing and Anjos. In Press). Brooding Plush- crested Jays were observed being ted by several individuals of the flock, as reported for most Cyanocorax jays (e.g.. Moore 1938. Crossin 1967). The type of vocalization (pair call) before feeding lias also been recorded for other Cyano- coiilx species: Curl-crested Jay (Amaral and Macedo 2003), Tufted Jay (C. dickeyi) (Moore 1938. Skutch 1987), and Azure Jay t Anjos and Vielliard 1993; Boesing and Anjos. In Press). Feeding of both nestlings and the brooding bird by helpers is common in several species of Cyanocorax jays (e.g.. Bosque and Molina 2002. Amaral and Macedo 2003). Breeders assisted by helpers were observed to fledge more young in the case of the Florida Scrub Jay (Aphelocoma coerulescens) (Woolfenden and Fitzpatrick 1984). a cooperative jay in North America. We observed possible predation of nestlings and fledglings by hawks (Roadside Hawk | Huteo magnirostris] and Yellow-headed C.aracara \Milvago chimachima]), marsupials (e.g.. Didelphis spp.), and reptiles (e.g.. Tupinambis spp.). One South American coati (Nasna nasua) was recorded preying on a fledgling on the ground. The most common cause of the lack of reproductive success in nests of corvids is predation (Marzluff 1985). Neotropical jays occupy different forest strata, but preferences vary among species (Anjos et al. 2009). The Plush-crested Jay frequents all lorest strata, but seems to prefer the understory and middle level of the forest, despite frequently using the ground, where food is available (VVP and KEP). All available layer-,, including the ground, are used by Inca Jays (Alvarez 1975) and Purplish-backed Jays (C. beecheii ) (Raitt and Hardy 1979). Azure Jays often only frequent the canopy of the forest (Anjos 1991). The availabil¬ ity of food at VVP favors use of resources on the ground, whereas the density of vegetation pro¬ motes the use of higher strata at RTF and KEP. Caching behaviors are widespread throughout the Corvidae. Food storage is commonly observed 94 THE WILSON JOURNAL OF ORNITHOLOGY • Voi 124. No. 1. March 2012 when birds have more food available than they are able or willing to eat at once, especially if the food is of a kind that will remain edible for some lime (Goodwin 1976). This behavior is poorly documented in Cyarutcomx jays. Caching behavior was observed in captivity by the Tufted Jay. but it does not appear to have been observed in the wild (Crossin 1967). Anjos (1991 ) observed this behavior in Azure Jays in both captivity and in the wild; this species was observed storing seeds of Araucaria crngustifolia in the canopy of the MF (Anjos 1991). The Flush- crested Jay caches the seed of A. angustifoliu on the ground, and it should be considered as an incidental disperser ol this tree species. Some seedlings of A. angustifolia were recorded in places, around the animal breeding center at KEP, where individual Plush-crested Jays were observed caching seeds of this tree species. The Plush-crested Jay exhibited variations in: ( I ) forest strata occupancy (preference for the under- story and middle level ol the forest, frequently using the ground where there was human food supplementation); (2) territory and home range sizes (both smaller when food supplementation was available); and (3) foraging behaviors (more diverse in less disturbed habitats). These charac¬ teristics demonstrate the plasticity of this species in habitat use, similar to most species of Corvidae. ACKNOWLEDGMENTS We are graietul tor logistical support in Vila Vclhn Stale Park. Klabin Ecological Park and Ribeirao do Tigre Farm. Personal of the Environmental Police of Parana State also assisted us during field work. Plants were identified by M. C. Dias (State University of Londrina). Financial support for field work was provided by the Federal University of Parana (UFPR>. Pedro Sherer-Neto, L. C. Millco, and F. C. Straube made important suggestions during the field work and revised an earlier version of the manuscript, which was presented as a Master Thesis at UFPR. We appreciate the valuable suggestions by J. M. Mar/luff. Tom Webber, and C. E. Braun, which improved the last version of this manuscript. The authors received research grants from CNPq (Brazilian Council for Development of Science and Technology) to the first and third authors, and from CAPES (Coordination for the Improvement of Higher Level Personnel. DS). to the second author. li i cKA I UKE CITED Altman* J. 1974. Oteryatoal ad, of beha, sampling methods. Behavior 49-227-^67 Amaral. M, F. and It 11 r Jay of central Brazil. Journal of Field Omitholo>i 74:331-340. ANJOS, L. 1 99 1, O ciclo anual de Cvcmocorax catrultw cm Floresta de Araucaria (Passeriformes: Corvidae Ararajuba 2:19-23. Anjos, I.. 2009. Family Corvidae. General introducin' Pages 494-565 in Handbook of the birds of the Volume 14 (J. del lloyo. A. F.llion. and D. A Clm i. Fditors), Lynx Editions, Barcelona. Spam. Anjos, L. and J. M. E. Vielliard. 1993. Repem.ric.d it. acoustic communication of the Azure Jay Cyannam « caeruleus (Viellot) (Aves. Corvtdaei. Revisu Bi leira de Zoologia 10:657-664. Anjos. l., s. Debus, S. Madge, and j. M Marzup 2009. Family Corv idae. Species accounts. Pages 5"'> 640 in Handbook of the birds of the world. Volutin (J. del Hoyu, A. Elliott, and D. A. Christie. Editor- Lynx Edit ions. Barcelona, Spain. BOKSINC. A. L. AND L. Anjos. In Press. The Azure Jay i. : reproduce in plantations of Araucaria angustifolui i" southern Brazil. Bird Conservation International BosguE, C. and C. Mol.ina. 2002. Communal brcedm; and nest defense behavior of the Cayenne (•" ( Cuirlocorax cayanus). Journal of l-ield Oniilhohvgv 73:360-36 2. Brunfj’ta. B. and L. Anjos. 2010. Variant* n) di.stribuiv&o espacial e no grito social da Gralha-p'.V (Cyanoconix chrvsnps) na Mata Atliinticu. sul * Brasil. Ornitologia Neotropical 21:203-213. C'ROSSIN. R. S. 1967. The breeding biology of the Tutterl Jay. Proceedings of the Western Foundation d Vertebrate Zoology 1:265-297. Dt.'CA, C.. T. J. GUF.RRA, AND M, A. MARINI. 2O0& Territory size of three antbirds (Aves, Passeriforme in an Atlantic Forest fragment in southeastern BraZ’1 Revista Brasileira de Zoologia 23:692-698. Fowler. J. and L. Cohen. 1986. Statistics for omithol. - gists. BTO Guide 22. British Trust for OmitfwkY I hetlord. United Kingdom. Gill. F. and D. Donsker (Editors). 2011. IOC World bit- names. Version 2.9. Princeton University Press, Prices ton. New Jersey. USA. hllp.V/w ww.woridbirdnamo org/ CiOODW in. D. 1976, Crows of the world. Cornell L'niversi .' Press. Ithaca. New York. USA. Hardy. J. W. 1961. Studies in behavior and phvJogeny 1 certain New World jays (Garrulinae). University 1 Kansas Science Bulletin 42:13-149. JULLIEN. M. AND J. T. ThiollaY. 1998. Multi-Sped^ territoriality and dy namic of neotropical forest urnk story bird (locks. Journal of Animal Ecology 6 - 252. LUGtNDUHL, J. M.. J. M. MARZLUFF, J. E. BRADLEY. M 0 Raphael, and D. e. Varland. 2001. Corvid Jimd techniques and the relationship between corvid rd- live abundance and nest predation. Journal ot F'L'U Ornithology 72:556-572. Madge. S. and H. Burn. 1994. Crows and jays: a guide- 1 the crows, jays and magpies of the world. Christoph*- Helm Ltd.. London. United Kingdom. Uejima et al • BREEDING AND FORAGING OF THE PLUSH-CRESTED JAY 95 Maklitf. J. M. 1985. Behavior at a Pinyon Jay nest in response to predation. Condor 87:559-561. NUniMT, j M. AND E. NfcATHE.Rl.iN. 2006. Corvid response to human settlements and campgrounds: rauwv consequences, and challenges for conservation. Biological Conservation 1 30: 301-314. v. oovex F. and I. M. Danni-Olivf.ira. 2002. Dinamia atmosferica e tipos climaticos predominates •la haciado Rio Tibagi. Pages 63-66 in A bad a do Rio Tibagi (M. E. Medri. E. Bianchini, O. A. Shihatta, and I A. Phnetiia, Editors). Londrina, Parana. Brazil. \Lviri Botanical Garden 2010. Missouri Botanical Garden plant list. St. Louis, Missouri, USA. www. impicos.org Mohr, C. 0, 1947. Table of equivalent populations of North American small mammals. American Midland Naturalist 37:223-249. Moore. R, T. 1938. Discover) of the nest and eggs of the Tutted Jay. Condor 40:233-241. Otttt. E. p. AND E. J. Kuf.N7.LER. 1955. Measurement of territory size and home range size in birds. Auk 72:128-137. Oliveira, Y. M. M. and E. Rotta. 1982. Levantamento da estrutura horizontal de uma mata de Araucaria do primeiro planalto paranaense. Bolctim de Pesquisa Florestal 4:M6. Raitt. R. J. and .1. w. Hardy. 1979. Social behavior, habitat, and food of the Bcechey Jay. Wilson Bulletin 91:1-15. Ribeiro. B. A., M. F. Goulart, and M. A. Marini. 2002. Aspect! is da terrilorialidade de Knipolegus lopltotes (Tyrannidae, Fluvicolinac.) em seu periodo reprodu- tivo. Ararajuba 10:23 1—235- RoBINSON. s. K. and R. T Holmes. 1982. Foraging behavior of forest birds: the relationships among search tactics, diet and habitat structure. Ecology 63:1918-1931. Silv \, J. M. C, AND J. M. Bates. 2002. Biogeographic patterns and conservation in the South American Cerrado: a tropical savanna hotspot. Bioscience 52:225-233. Ski TCH. A. F. 1987. Helpers at birds' nests. A worldwide survey of cooperative breeding and related behavior. University of Iowa Press. Iowa City, USA. WlKTANDER. U., O. OLSON. AND S. G NILSSON. 2001. Seasonal variation in home-range size, and habitat area requirement of the Lesser Spotted Woodpecker (Dendrocnpos minor) in southern Sweden. Biological Conservation 100:387-395. WOOLEENDEN, G. E. AND J. W. FITZPATRICK. 1984. The Florida Scrub Jay: demography of a cooperative¬ breeding bird. Princeton University Press, Princeton, New Jersey, USA. The Wilson Journal of Ornithology 124(1):96-105, 2012 SEXUAL SELECTION AND MATING CHRONOLOGY OF LESSER PRAIRIE-CHICKENS ADAM C. BEHNEY,157 BLAKE A. GRISHAM.' CLINT W. BOAL,* 2 * HEATHER A. WHITE AW.' 6 7 AND DAVID A. HAUKO 5S4 ABSTRACT. — Little is known about mate selection and lek dynamics of Lesser Prairie-Chickens ( Tympanuchus pallidicinctus). We collected data on male territory size and location on leks. behavior, and morphological characteristics and assessed the importance of these variables on male Lesser Prairie-Chicken mating success during spring 2008 and 2009 in the Texas Southern High Plains. Wc used discrete choice models and found that males that were less idle were chosen more often for mating. Our results also suggest that males with smaller territories obtained more copulations. Morphological characteristics were weaker predictors of male mating success. Peak female attendance at leks occurred during the 1 -week interval starting 13 April during both years of study. Male prairie-chickens appear to make exploratory movements to. and from, leks early in the lekking season; 13 of 19 males banded early (23 Feb-13 Mar) in the lekking season departed the lek of capture and were not reobserved ( I I yearlings, 2 adults). Thirty-three percent (range = 26-51' h of males on a lek mated (yearlings = 44%. adults - 20%) and males that were more active experienced greater mating success. Received 2 May 2 Oil. Accepted 2S July 2 Oil. Males in lek mating systems aggregate on arenas (leks) which females visit for breeding; males provide no parental care or resources to females, other than genetic material (Hoglund and Alatalo 1995). Sexual selection is typically strong in lek mating systems where some individuals obtain many mating opportunities while others obtain none (Robel 1966, Gibson and Bradbury 1985, McDonald 1989) and, in many species, males have evolved elaborate courtship displays and ornaments. Females are thought to select the highest quality males to maximize direct (survival or clutch size.) or indirect benefits (good genes) (Bradbury and Gibson 1983, Reynolds and Gross 1990). Vocal, morphological, territorial, and behavior¬ al characteristics have been examined among lekking grouse species with regard to mate choice (e.g., Robel 1966. Gibson and Bradbury 1985, 'Texas Cooperative Fish and Wildlife Research Unit, Department of Natural Resources Management. Texas Tech University. Lubbock. TX 79409, USA. 2 U.S. Geological Survey, Texas Cooperative Fish and Wildlife Research Unit. Department of Natural Resources Management, Texas Tech University. Lubbock, TX 79409, USA. 'Texas Parks and Wildlife Department, Texas Tech University, Lubbock, TX 79409, USA. U.S. f ish and Wildlife Service, Texas Tech University Lubbock, TX 79409, USA. 5 Current address: Cooperative Wildlife Research Labo¬ ratory. Southern Illinois University, Carbondale. IL 62901 USA. "Current address: U.S. Fish and Wildlife Service. T< tech University, Lubbock. TX 79409. USA 7 Corresponding author; e-mail; abehney@siu edu Gibson et al. 1991, Gibson 1996. Hoglund et al. 1997, Nooker and Sandercock 2008). Correlates of male mating status (mated vs. non-mated) for Greater Sage-Grouse ( Centrocercus urophasia- nus) included display rate, lek attendance, and a vocal component (Gibson and Bradbury 1985). Gibson cl al. (1991) found that female choice in Greater Sage-Grouse was related to male vocal¬ ization performance, previous mating locations o? females, and choices of other females. Specifical¬ ly, initial attraction of female Greater Sage- Grouse to males was based on vocalizations while probability of mating was related to male display rate (Gibson 1996). Male Sharp-tailed Grouse ( Tympanuchus phasianellus) holding central ter¬ ritories obtained more copulations than peripheral males (Gratson et al. 1991), although Gratson (1993) concluded that dance time and auditory characteristics were better predictors of matin;: success than territory' location. Alternatively, display and aggressive behaviors were better predictors of male mating success for Greater Prairie-Chickens (T. cupido) than territory char¬ acteristics (Nooker and Sandercock 2008). Lesser Prairie-Chickens (T. pallidicinctus) are a lek-mating gaxise. inhabiting short and mixed grass prairies of the southern Great Plains. Signit icant population declines throughout much of l her historic range (Hagen and Giesen 2005) have resulted in their designation as a 'candidate for protection under the Endangered Species Ad (USDI 2008). Little is known about sexual selection and lek dynamics of Lesser Prairie-Chickens, and future research and conservation could benefit I rod information on when prairie-chickens mate, how 96 Behneyetal. • PRAIRIE-CHICKEN BREEDING BEHAVIOR 97 many males mate, and what characteristics influ¬ ence male mating success. The objectives ot our study were to (1.) assess the roles of behavioral, territorial, and morphological characteristics tor Lesser Prairie-Chicken mate choice, (2) report dates if peak female attendance and copulations, and (3) issess the extent of mating skew on prairie-chicken Ids. METHODS S ntd\ Area,— Our study occurred on private itos in Cochran and Yoakum counties in the lews Southern High Plains Ecoregion (Llano Esiacado). The area consists of a matrix of grassland and cropland (Wu et al. 2001) among j level to gently undulating landscape with small ■egetated dunes providing infrequent topograph¬ 'll relief. The dominant vegetation was shinnery to iQuercus havardii ) intermixed with sand ■agebnish ( Artemisia filifolia), grasses, and forbs 'Pettit 1979, Woodward et al. 2001 ). ihe mean annual precipitation was 48.3 cm for to period 2000-2009 (50.3 and 45.2 cm in 2008 ■to 2009. respectively) with average summer Tun-Aug) and winter (Dec-Feb) temperatures ’I 35.4 and 5.4 C, respectively. Extreme high to low temperatures were 39.5 and -13.4 C, especti vely (U.S. Department of Commerce -1 h) The average elevation of the study area is '"U00 m. Held Methods,— "We conducted this study on |,l||r different leks during spring 2008 and 2009. I " ° leks were sampled in 2008 and three in 2009 ""h °ne sampled in both years. Grass cover on rv leks was too high and dense to see the birds Ts continuously. Thus, we selected leks lor this 'toy based on vegetation characteristics that 1 ditated identification of color bands on legs ot I^aine-chickens. 'Ve captured male Lesser Prairie-Chickens walk-in-funnel traps (l laukos ct al. 1990. Boeder and Braun 1991) early in the lekking eas,|n date Feb-early Mar). We also captured males 1 opportunistically with a bownct throughout * lcLking season. We did not attempt to capture iales with the bownet while females were present Wi. Each captured male was fitted with a ,u4ue color band combination (Association ot lC'd Ornithologists, Manomet, MA. USA) and a tobered aluminum Texas Parks and Wildlife bailment band (size 6). We measured mass (g). Sh pinnae length (mm), right tarsus length nni), and right unflattened wing cord length from bend of wing to tip of longest primary (mm) for each captured male. We classified prairie-chicken age as either adult or yearling based on plumage characteristics. Yearlings exhibited frayed tips of the ninth and tenth primaries and spotting within 2.5 cm of the tip of the tenth primary whereas adults lacked frayed primaries and had no spotting within 2.5 cm of the tip of the tenth primary (Copelin 1963). Four males (2 yearlings, 2 adults) were marked with necklace style radio transmitters. We placed a grid of points centered on the activity center of each study lek to facilitate mapping of male territories. Grid points consisted of numbered, orange-colored, blocks ot wood (7.6 X 5 1 X 5,1 cm), placed every 5 m encompassing the entire lek area. Some leks were sufficiently small to be covered with a 5 X 8 grid (20 X 35 m) while others required a 10 X 10 grid (45 X 45 m). Grids were placed on leks in February before birds started attending leks. We conducted observations from a blind (Primos Ground Max. Flora, MS, USA.) placed within 10 m of the edge of the lek during morning and evening lekking periods. We used binoculars and spotting scopes to identify males, and assess locations and behavior. We used the grid points as a reference to plot locations ot males onto a corresponding paper copy of the grid during 10- min interval scan samples. Lek observations were not conducted if a lek had walk-m-iunnel traps present or after the bownet had been triggered Observations were made 2-3 days/week Irom 2 February to 21 May 2008 and 5 March to 10 May o()09 The order of leks to be monitored was randomly selected, weekly. Leks were not ob¬ served when lightning was present or winds exceeded ~45 km/hr. . We recorded a description of male Lesser Prairie- Chicken behavior every time a location was plotted. Behavioral categories included display, moving face off. fighting, and idle. Display involved erecting pinnae, enlarging eye-combs, elevating tail drooping wings, extending head and neck forward, stamping feet, inflating esophageal air sacs, and emitting booming vocalization (Hagen and Giesen 2005). Moving was when the male was walking or running but not displaying. Face off consisted of two males in close proximity (<1 m), facing each other at a territory boundary. lypically in a semiprone position, but not displaying, moving, or fighting (Hagen and Giesen 2005). Fighting consisted of two males actively fighting each other 98 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 with one typically charging the other with rapid aggressive movements. Idle was recorded when a male was not doing any of the other behaviors. We noted the male that performed any copulation attempt and the location. A copulation attempt was defined as anytime a male was able to put at least one foot on the female’s back. Most copulation attempts were interrupted by other males at varying times throughout the attempt. We classified copulation attempts as successful when females vigorously ruffled their feathers and departed the lek after copulating (Hagen and Giesen 2005). Statistical Analyses— We used .Skew Calcula¬ tor 2003 (Nonacs 2003) to analyze mating skew on leks. We used two indices of mating skew: X and R (Binomial Skew Index). Lambda values ranged from 0 to I with larger values indicating greater skew (Kokko and Lindstrom 1997). Positive values of R indicate some monopoliza¬ tion ot matings (skew), while 0 indicated random mating, and negative values indicated a more equal distribution (Nonacs 2000). /i-values, gen¬ erally. ranged from - 1 to I. although it is possible to obtain values >1. Confidence intervals and P values can be calculated for R (Nonacs 2000). Lek attendance rates were calculated as the number of days a male was observed on the lek divided by the number ol days the lek was observed and at least one male was present. We assessed two characteristics of territories: size and distance to center of lek activity. We calculated two measures of territory size, 95% kernel and 95% minimum convex polygon (MCP) estimates, using plotted male locations with >24 locations/behavioral observations. Both metrics were computed in the ADEHABI- TAT package (Calenge 2006) of Program R (R Development Core Team 2008). Lesser Prairie- Chickens appear to spend a disproportionate amount of time at territory boundaries and we suspect that kernel estimators overestimated territory size. MCPs only outline the outer points of a distribution and may be more accurate in assessing individual territory sizes of ickking prairie-chickens. Thus, only MCP estimates of territory size were used Tor modeling. We report ernel estimates of territory size for comparative puiposas with other studies of lekking grouse ( g„ Nooker and Sandcrcock 2008). Kernel (r =aoo«Te a "ed With MCP est'mates , • )' A ma,c s center of activity was computed as the centroid of all its locations Each male’s centroid was averaged to ascertain the center of activity for the lek. Discrete choice models (DCM) allow inference to be drawn about resource preferences based on the attributes of the resource (Cooper and Mill- spaugh 1999). These models predict the probabil¬ ity that an individual will select a certain resource as opposed to any of the other available resources and assume that individuals make choices that will maximize utility (Cooper and Millspaugh 1999). DCMs are used more frequently in habitat selection studies (e.g., Lesmeister et al. 2008. Vanak and Gompper 2010). We followed the example of Nooker and Sandercock's (2008) Studies of Greater Prairie-Chickens and used DCMs (PROC MDC, SAS Version 9.1. Can', NC, USA) to assess correlates of male mating success for Lesser Prairie-Chickens. Each copu¬ lation attempt represents one sample in the DCM A female chooses one male to mate with among a group ol males, which is considered the choice set. DCMs allow the choice set to vary by sample, which is necessary when multiple leks are involved. The males (or sample of males) on one lek compose the choice set for each copulation attempt on that lek. We had to colled >24 location/behavior points on the male in¬ volved in the copulation attempt for it to be included in this analysis. Not every male on a Id was included in the analysis, but we believe our sample is representative of all males attending the lek. We trapped across the entire lek area and did not focus trapping efforts on central or peripheral males. Behavioral variables included the proportion ol observations recorded as each behavior category display, face off, fighting, idle, and moving Morphological variables included wing cord length (cm), tarsus length (cm), pinnae length (cm), and mass (g). Territorial variables included distant to lek center (m), and territory size (MCP. m We did not use all variables in discrete choice models due to small sample sizes. We selected the behavioral variables display and idle to use w models because they were uncorrelated (r = "9.1, P — 0.59) and represented what we hypothesized to be important in mate choice. We selected the morphological variables mass and pinnae length because they were uncorrelated (/• = 0.1 ,P~ 0.591 and represent a size component (mass) and a secondary sexual characteristic (pinnae). We ah'" included territory size, distance to lek center, and age in models. Belviev et al. • PRAIRIE-CHICKEN BREEDING BEHAVIOR 99 Variables were standardized by replacing each observation by ixrx^/sXi for each lek to facilitate dual comparison of parameter estimates as effect AIK (Gnison et al. 1991. Agresti 2002. Nooker and Sandercock 2008). The sign of the slope i 'efficients indicate if that variable is positively • negatively correlated with male mating success and the magnitude of coefficients are directly comparable indices of effect size. We used \kaike’s Information Criterion corrected lor small -amples I AIC( ) and model averaged slope coefficient estimates across all models in the model set to avoid basing inference on a single model (Anderson 2008). We only considered models with ^3 variables due to small sample sizes. Each variable appeared an equal number of times in the model set to facilitate model averaging and calculating relative importance values (Anderson 2008). We evaluat¬ 'd models based on all copulation attempts regardless of whether it was successful, and only successful copulation attempts. We also calculat¬ ed Pearson’s correlation coefficient (r) between each variable and the proportion of all and 'Uccessful copulations each male obtained on >ls’ respective lek. is generally not good practice to use all possible models but we believe il was justified due lo the exploratory nature of this type of analysis. Previous studies have not examined e*ua! selection of Lesser Prairie-Chickens and °Ur goal was to provide a baseline for more in- kp'h future experimental work. All models were biologically and theoretically possible and we USl'd model averaging to derive parameter esti ina!es as indices of effect sizes so inference was 1,111 placed on any single model (Anderson el al. 20f|0. Anderson 2008). RESULTS We spent 272.5 hrs observing Lesser Prairie- f-hicken behavior at leks during spring 2008 and (mean ± SE = 47.9 ± 6.9 hrs/lek/yr). Study averaged 10.5 males/morning (range = 4.4- during the spring lekking season. Wc Captured 22 and 14 birds in 2008 and 2009. Actively. Thirteen of nineteen males (M yearlings, 2 adults) captured during early trapping essions between 23 February and 1 3 March 2008, Were not reobserved even after extensively Aching within 4 km of the leks of capture. Tto*se 13 birds were not included in the analyses. Additionally, two males in 2008 and 2009 were banded but we were unable to collect all morphological measurements. This left us with 7 and 12 individuals, respectively, in 2008 and 2009 with complete morphological measurements to use for analysis. Mean ± SE lek attendance rate of marked males that were reobserved on study leks at least once was 0.88 ± 0.04. We noted 163 and 76 female observations on leks in 2008 and 2009, respectively. Female lek attendance peaked during the 7-day interval starting 13 April in both years (Fig. 1A). The maximum number of females observed on a lek simultaneously was 17. We observed females on leks during evening display periods on one and two occasions in 2008 and 2009, respectively, and in 2008 we observed toui copulation attempts during evening lekking. Overall, male mating success was skewed (X - 0.60; 5-valuc = 0.30. P < 0.001). We observed 62 copulation attempts on leks in 2008. 30 ot which were deemed successful. Copulation attempts peaked during the 7-day interval starting 27 April (Fig. IB). Four males were responsible for all copulation attempts on lek Bl, which averaged 15.2 males per morning!/. = 0.54; 5-value - 0.27. P < 0.001). Three males were responsible for 97% of copulation attempts, two ol which were responsible for 82% of all copulation attempts. Five males were responsible for all copulation attempts on lek B2, which averaged 16.0 males per morning (X = 0.77; /7-value = 0.53. P < 0.001 . Three males were responsible tor 93 ’b of al copulation attempts, one of which performed 79% of all copulation attempts. We observed 2) copulation attempts in 2009, 12 of which were deemed successful. Copulation attempts peaked during the interval starting 13 April (Fig. I B). We only observed one copulation attempt on lek B , and this lek was removed from the skew analysis. Three males were responsible for all copulation attempts on lek B4, which averaged 9.2 males per morning (X = 0.58; 5-value = 0.24, P -- 0.001). Two males were responsible for 88% of all copulation attempts. Four males performed all copulation attempts on lek R5, which averaged 7.8 males per morning ( X = 0.52; 5-value = 0.15, P = 0.006) with two of the males performing 82% of all copulation attempts. The percentage of adult and yearling marked birds that attempted ^1 copula¬ tion w'as 20 and 44%, respectively. The mean ± SE percentage of copulations obtained on a lek for adults and yearlings was 0.09 ± 0.06 and 0.18 ± 0.05, respectively. 100 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 9 JS' !tTle Tdr 0.1) with proportion of copulations obtained The most parsimonious model, considering ail copulation attempts (n = 52 copulation attempts- 19 males), included percent of time spent idle anct MCP, and obtained 39% of the Akaike weigh1 (Table 2). All models containing idle were tanked higher than those not containing idle, which also had the greatest model averaged slope coefficient (Fig. 3). The third and fourth best model1- appeared competitive but contained pretending variables, which contribute little to model lit as evidenced by similar deviance values. Relame importance values for idle. MCP. age, pinnae, display, distance, and mass were 1.00. 0.52. 04 0. J 3, 0. 1 2. 0.00, and 0.00. respectively. T he most parsimonious model, considering only successful copulation attempts (n = 30 copulations. Behneyetal. • PRAIRIE-CHICKEN BREEDING BEHAVIOR 101 TABLE 1. Characteristics measured for male Lesser :0()9 in ihe Texas Southern High Plains. Prairie-Chickens observed at two leks in 2008 and three leks in Mean i SE Calfjorv Trail Adult in = 10) Yearling ( n = 9) CV Behavior6 Territory Morphology Display Face off Fighting Idle Moving Distance to lek center (m) Kernel size (nr) MCP size (nr) Wing cord (cm) Tarsus (cm) Pinnae (cm) Mass (g) 0.29 ± 0.02 0.33 ± 0.03 0.00 ± o.oo 0.25 ± 0.03 0.13 ± 0.02 12.66 ± 1.40 245.07 ±41.26 108.50 ± 17.49 21.69 ± 0.12 5.83 ± 0.06 6.64 ± 0.13 783.10 ± 9.03 0.30 ± 0.03 0.35 ± 0.04 0.01 ± 0.00 0.19 ± 0.02 0.15 ± 0.01 1 1.09 ± 1.75 109.68 ± 10.73 43.29 ± 5.46 21.54 ± 0.09 5.47 ±0.11 6.88 ± 0.22 780.22 ± 13.65 36.57 41.63 295.55 53.36 44.64 56.44 81.92 84.27 2.17 7.42 11.08 6.20 h Coefncicnl of Variation values computed from pooled adult and yearling values. ’ |W»w variables are proportion of observations in each behavior category for nn individual male. males), included idle and MCP hut obtained an Akaike weight of only 23% (Table 3). All models containing idle outperformed those not containing ihe variable and it had the greatest model averaged slope coefficient (Fig. 3). Relative importance values for idle, MCP, age, mass, pinnae, display, and distance were 1.00, 0.43. 0.39. 0.27, 0.17,0.08, and 0.05. respectively. DISCUSSION found significant skew in male mating, S|milar to those reported for Greater Prairie- chickens (Nooker and Sandercock 2008). It is tlear (hat male Lesser Prairie-Chickens that are less idle experience greater mating success, darling males with smaller territories also tended ’"k selected more often for mating in our study. Morphological characteristics exhibited weaker effects on male mating success. Males displayed high territory fidelity within a season (alter initial territory establishment). Males that were less idle were more likely to mate which we interpret to indicate that males that are generally more active experience greater mating success. It has been repeatedly tound that males that display more, mate more (Gibson and Bradbury 1985. Hdglund and Lundberg 1987, Nooker and Sandercock 2008). Being idle likely requires less energy than participating in other behaviors and Gibson and Bradbury (1985) suggest that energetic factors may have a role in observed variation in display rates. A host ot reasons exist for female choice based on behav¬ ioral characteristics including direct survival benefits for the female or indirect genetic benefits for her offspring. For example, males that do not 2. Pearson’s correlation coefficient ^^^."^Ton let-in meTex^s'stu'themffigh Plains during 2008 and 4. Splay' JIS Me. aTmoving are ,he proportion of behaviors recorded in each behavior category. Pittance = distance from territory center to lek center. MCP = territory size tmmtntnm convex polygon). Wmg, tarsus, P'nnae. and mass are morphological characteristics. 102 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. 1. March 2012 TABLE 2. Top (A < 20) conditional logit discrete choice models of male Lesser Prairie-Chicken mating success in the Texas Southern High Plains during 2008 and 2009 incorporating 19 males and 52 copulations regardless of success. Model' Deviance Idle + MCP Idle + Age Idle + MCP + Pinnae Idle + Age + Display Idle + Dist + Mass Idle + Pinnae Idle + Mass Idle + Dist + Pinnae Idle AIC, AAIC, w , 60.95 61.17 60.95 61.14 72.80 75.58 77.32 75.21 81.47 65.17 65.38 67.40 67.58 79.24 79.79 81.53 81.66 83.54 0.00 0.21 2.22 2.41 14.07 14.62 16.36 16.48 18.37 0.40 0.36 0.13 0.12 0.00 0.00 0.00 0.00 0.00 fiESSSe 2 227-3222222125 SUES!" “ — «» ■ — “ display often may indicate a poorer physiological condition and inability to acquire sufficient "food resources compared to other males. Males with smaller territories tended to mate more, as reported by others (Wiley 1973, Hovi et al. 1994). Our finding of distance to lek center having little to no effect on mating success is in contrast to most previous research (Ballard and Robel 1974, Kruijt and de Vos 1988, Gratson et al. 1991, Rintamaki el al. 1995) although Gibson and Bradbury (1985) and Nooker and Sandercock (2008) also found that territory location was not important in mate choice. Our correlation analysis suggested that males closer to the center of the lek mated more than peripheral males. Smaller terri¬ tories are typically associated with areas on the lek with higher male density (Wiley 1973). Areas of high male density arc thought to be a result of males relocating their territories around successful males and intruding into their territories in hope of gaining copulations (Landel 1 989). Rintamaki et al. (1995) noted this phenomenon as the ‘spatial spill* hypothesis (hotshot hypothesis, Arak 1984), whereas males cluster around dominant males in hope of gaining copulations. Rintamaki et al. (1995) speculated the reason for these ‘spillover copulations may include a surplus of females attempting to copulate with the dominant male and competition lor that male may cause females to mate with adjacent males. The dominant male may be limited by sperm depletion or adjacent males may steal copulations from a preoccupied domi¬ nant male (Rintamaki et al. 1995). Females may experience difficulties in comparing males and mistakenly mate with an adjacent, potentially poorer quality, male. It is not clear whether territory size or location is a cause or effect of being a dominant male (Gratson et al. 1991). It is possible that radio transmitters affected reproductive performance of prairie-chickens. The lour radio-marked males were all on the same lek and included two adults and two yearlings. Only variable choice models de^cribimf effwTV C°efru;ients (P,US or m,nus unconditional SE) of standardized variables from discrete (light gray) and only successful c T c Le^er Prairic'Ch,cken characteristics on obtainment of any copulation attempt categorical adult or yearling.' DisoUv miTdl (dark.gray) ,n ,he Texas Southern High Plains during 2008 and 2009. Age is is distance from territory center lo lek - ^ proponjon nl behaviors recorded as each behavior category. Distance morphological characteristics Frm.- hirc^ ls ,erntory size (minimum convex polygon). Pinnae and mass are • trro, bars extending outs.de the region 5 to -7 are not shown completely. Behneyetal. • PRAIRIE-CHICKEN BREEDING BEHAVIOR 103 TABLE 3. Top (A < 10) conditional logit discrete choice models of male Lesser Prairie-Chicken mating success in the Texas Southern High Plains during 2008 and 2009 incorporating 19 males and 30 successful copulations. Model- k Deviance AICf AA1C, W/ Idle - MCP 2 22.96 27.41 0.00 0.24 Idle + Ase 2 23.10 27.54 0.13 0.22 Idle - MCP + Mass 3 22.06 28.98 1.57 0.11 Idle + Ase - Mass 3 22.34 29.27 1.86 0.09 Idle - MCP + Pinnae 3 22.58 29.50 2.10 0.08 Idle + Age + Display 3 23.01 29.93 2.52 0.07 Idle + Pinnae 2 25.60 30.04 2.63 0.06 Idle + Mass 2 26.35 30.80 3.39 0.04 Idle + Dist + Mass 3 25.15 32.08 4.67 0.02 Idle - Dist + Pinnae 3 25.59 32.5 1 5.11 0.02 Idle 1 30.64 32.78 5.38 0.02 Idle - Display 2 29.63 34.08 6.67 0.01 Idle + Dist 2 30.58 35.02 7.62 0.01 ' Ule = proportion of time spent idle; Display = proportion of time spent displaying; Dist - distance from territory center to lek center. MCP territory size emmum convex polygon); Pinnae and Mass arc morphological measurements; Age = yearling or adult. one of Ihe radio-marked males mated (a yearling). Lillie information is available on the effects of radio transmitters on male grouse reproductive performance although Boag (1972) reported radio-marked captive Red Grouse ( Lagopus lagupus scotica ) were less active than controls. The small sample of radio-marked males in our 9udy prevented any test of effects. Territory occupancy stabilized —13 March in -MK. Males captured after 1 3 March during both years of study were reobserved on the lek of capture whereas in 2008, many males captured Ndore 13 March were not reobserved. In contrast. Haukos (1988) reported that, within the same area, territories were unstable and he did not 1 i,serve any copulation attempts on leks whereas "e observed 91. We suspect this inconsistency ma.v he due to differences in vegetation on leks belvveen the two studies. Haukos (1988) reported fetation on the leks was sparse, if present at all. and physical structure was frequently altered by 'ind. Our study occurred 20 years later and accession of vegetation had covered leks with Sh°rtgrass and small shrub cover. banded numerous males on leks early in the -kking season that did not establish territories at '•se sites. Unfortunately, we did not radiomark *** individuals. Thus, fate of the males that ^parted their lek of capture and were not observed is unknown. Hagen et al. (2005) lound •to some yearling (20%) and adult (8%) males 'Witched leks between years with an average ^ce traveled of 3.3 and 3.1 km. respectively, was well within our search areas but we failed lo relocate any of the males at other leks. Haukos and Smith (1999) observed similar patterns of male movements and satellite lek formation just prior to female attendance. Our data, and those of Haukos and Smith (1999) and Hagen et al. (2005) suggest that estimates of population size from lek counts may be biased if these males did not establish territories on a lek (Walsh ct al. 2010). Research using early season radiomarking to examine early season dispersal and inter-lek movements within a season could prove valuable for understanding lek dynamics and gene flow, and facilitate bettei estimates of population size, ACKNOWLEDGMENTS Funding was provided by the Texas Parks and Vvildlite Department. The Department of Natural Resources Man¬ agement at Texas Tech University and the USGS Texas Cooperative Fish and Wildlife Research Unit also contrib¬ uted resources. This research would not have been possible without the private landowners who graciously allowed us to work on their land Wc thank D R. Lucia for assistance throughout this research as well as N. E. Pinus, A. J. Teague. C G. Frey. R. T. Sadowski. and C J. Kveton lor help in the field. Wc also thank D. B Lesmeister for assistance with discrete choice models. M I Butler. P. B. Wood. C. A. Hagen. R. M. Gibson, and two anonymous reviewers provided valuable comments on this manuscript. T rade name products are mentioned to provide complete descriptions of methods: the author's and their institutions neither endorse these products nor intend to discriminate against products not mentioned. LITERATURE CITED ACiRESTl, A. 2002. Categorical data analysis. John Wiley and Sons, Hoboken, New Jersey. USA. 104 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. I. March 2012 Anderson, D. R. 2008. Model based inference in the life sciences. Springer, New York, USA. Anderson, D. R„ K. P. Burnham, and W. L. Thompson. 2000. Null hypothesis testing: problems, prevalence, and an alternative. Journal of Wildlife Management 64:912-923. Arak, A. 1984. Sneaky breeders. Pages 154-194 in Producers and scroungers: strategies of exploitation and parasitism (C. J. Barnard. Editor). Croom Helm, Beckenham. United Kingdom. Ballard, W. B. and R. J. Robel. 1974. Reproductive importance of dominant male Greater Prairie Chick¬ ens. Auk 91:75-85. Boag, D. A. 1972. Effect of radio packages on behavior of captive Red Grouse. Journal of Wildlife Management 36:511-518. Bradbury. J. W. and R. M. Gibson. 1983. LekS and mate choice. Pages 109-138 in Mate choice (P. Bateson. Editor). Cambridge University Press, New York, USA. Calenge, C. 2006. The package ADEHABITAT for the R software: a tool for the analysis of space and habitat use by animals. Ecological Modeling 197:516-519. Cooper, A. B. and J. .1. Millspaugii. 1999. The application of discrete choice models to wildlife resource selection studies. Ecology 80:566-575. Copelin, F. F. 1963. The Lesser Prairie Chicken in Oklahoma. Technical Bulletin Number 6. Oklahoma Wildlife Conservation Department, Oklahoma City, USA. Gibson, R. M. 1996. Female choice in Sage Grouse: the roles of attraction and active comparison. Behavioral Ecology and Sociobiology 39:55-59. Gibson. R. M, and J. W. Bradbury. 1985. Sexual selection in lekking Sage Grouse: phenotypic corre¬ lates of male mating success. Behavioral Ecology and Sociobiology 18:117-123. Gibson, R M.. J. W. Bradbury, and S. L. Vehrencamp. 1991. Mate choice in lekking Sage Grouse revisited: the roles of vocal display, female site fidelity, and copying. Behavioral Ecology 2:165-180. Gratson. M. W. 1993. Sexual selection for increased male courtship and acoustic signals and against large male size at Sharp-tailed Grouse leks. Evolution 47:691- 696. Gratson. M. W„ G. K. Gratson, and a. T. Blrgerud. 1991. Male dominance and copulation disruption do not explain variance in male mating success on Sharp¬ tailed Grouse (Tympanuchus phasianellus) leks. Be¬ havior 118:187-213. Hagen, C. A. and K. M. GtESEN. 2005. Lesser Prairie- Chicken ( Tympanuchus pallidicmctus). The birds of North America, Number 364. Hagen, C. a., J. C. Pitman. B, K. Sandercock, R. .1. Robel. and R. d. Applegate. 2005. Age-specific variation in apparent survival rates of male Lesser Prairie-Chickens. CondOr 107:78-86. Haukos. D. A. 1988. Reproductive ecology of Lesser Praine Chickens in West Texas. Thesis. Texas Tech University, Lubbock. USA. Haukos, D. A. and L. M. Smith. 1999. Effects of lek age on age structure and uttcndunce of Lesser Prairie-Chickens ( Tympanuchus pallidicinctus). American Midland Nat¬ uralist 142:415-420. Haukos. D. A.. L. M. Smith, and G. S. Broim. 1990 Spring trapping of Lesser Prairie Chickens. Journal of Field Ornithology 61:20-25. HOGI t nd, J. AND A. Lundberg. 1987. Sexual selection in j monomorphic lek-breeding bird: correlates of male mating success in the Great Snipe Gallinagn media. Behavioral Ecology and Sociobiology 21:211-216. HOGLLND. J. and R. V. ALATALO. 1995. Leks Pnncctnn University Press. Princeton, New Jersey. USA HOglund. J., T Johansson, and C. Pelabon. 1997. Behaviourally mediated sexual selection: charactcns- tics of successful male Black Grouse. Animal Behaviour 54:255-264. Hovi, M„ R. V. ALATALO, J. Hoclund. A. Lundblro. and P. T. Rintamaki. 1994. Lek centre attracts Black Grouse females. Proceedings of the Royal Society of London. Scries B 258:303-305. Kokko, H. and J. LiNDSTROM. 1997. Measuring the mating skew. American Naturalist 149:794-799. Kkltjt. J. P. and G. J. or Vos. 1988. Individual variation in reproductive success in male Black Grouse, Teinm tutrix 1. Pages 279—290 in Reproductive success (T. H Clutton-Brock. Editor). University of Chicago Press. Chicago, Illinois, USA. Landhl, II. F. 1989. A study of female and male mating behavior and female mate choice in ihc Sharp-tailed Grouse. Tympanuchus phasianellus jamesi. Disserta¬ tion. Purdue University. West Lafayette. USA. LeSMEISTBR. D. B.. M. E. GoMPPER. AND J. J. MlLLSPAtCH 2008. Summer resting and den site selection by easlem spotted skunks (Spilogale putorius) in Arkansas. Journal of Mammalogy 89:1512-1520. McDonald, D. B 1989. Correlates of male mating success in a lekking bird with male-male cooperation. Animal Behaviour 37:1007-1022. NonaCS, P. 2(XK). Measuring and using skew in the study of social behavior and evolution. American Naturalist 156:577-589. NONACS. P. 2003. Skew calculator 2003. University of California. Los Angeles. USA. http://www.eeb uda. cdu/Faculty/Nonacs/share vvare.htm Nooker. J. K AND B. K. Sandercock. 2008. Phenotypic correlates and survival consequences of male mating success in lek-mating Greater Prairie-Chickens (Tvm- panuclius cupido). Behavioral Ecology and Sociobiol- ogy 62:1377-1388. Pettit, R. D. 1979. Effects of picloram and tehuthiuron pellets on sand shinnery oak communities. Journal of Range Management 32:196-200. R Development Core Team. 2008. R: a language and environment for statistical computing. R Foundation for Statistical Computing. Vienna. Austria. Reynolds. J. D. and M. R. Gross. 1990. Costs and benefits of female mate choice: is there a lek paradox" American Naturalist 136:230-243. Rintamaki. P. T„ R. V. Alatalo, J. HOglund. and A. LUNDBERG. 1 995. Male territoriality and female choice on Black Grouse leks. Animal Behavior 49:759-767. Behneyetal. • PRAIRIE-CHICKEN BREEDING BEHAVIOR 105 ROBa, R. J. 1966. Booming territory size and mating success of the Greater Prairie Chicken ( Tympanu - chut cupido pinnatus). Animal Behaviour 14:328- 331. SantOEDER. M. A. and C. E. Braun. 1991. Walk-in traps lor capturing Greater Prairie Chickens on leks. Journal of field Ornithology 62:378-385. i s Department of Commerce. 2010. Annual climato¬ logical summiry. National Climatic Data Center. National Oceanic and Atmospheric Administration. Washington. D.C.. USA http://cdo.ncdc.noaa.gov/ uncsum/ACS?eoban =41 6074 U.S. Department of Interior (USDI). 2008. Endangered and threatened wildlife and plants; review of native species that are candidates for listing as endangered or threatened. Federal Register 73:75176-75244. Vanak. A. T. and M. E. Gompper. 2010. Multi-scale resource selection and spatial ecology of the Indian fox in a human-dominated dry grassland ecosystem. Journal of Zoology 281:140-148. Wal.su, d. P.. J. R. Stiver, (i. C. White. T. E. Remington, and A. D. Apa. 2010. Population estimation techniques for lekking species. Journal of Wildlife Management 74:1607-1613. Wiu-.Y. R. 11. 1073. Territorial ily and noil-random mating in Sage Grouse, Cent race reus urophasianus. Animal Behavior Monographs 6:85-169. Woodward. A. J. W.. S. D. Puhlendorf, D. M. Leslie Jr., and J. Shackford. 2001. Influence of landscape composition and change on Lesser Prairie-Chicken ( Tympanuchus pallidicinctus ) populations. American Midland Naturalist 145:261-274. Wu, X. B„ N. J. Silvy, F. E. Smeins, and R. C. Maggio. 2001. Landscape changes in Lesser Prairie-Chicken habitat in the Texas panhandle. Report to Texas Parks and Wildlife Department, Austin. USA. The Wilson Journal of Ornithology I24( 1): 106-1 12, 2012 LEK BEHAVIOR OF THE PLOVERCREST (STEPHANOXIS LALANDI , TROCHILJDAE) MARCO AURELIO PIZO1 ABSTRACT.—! examined the lek structure and behavior of male Plovercrests ( Stephanoxis lalandi ) at a lek in southern Brazil. The lek included seven territorial males; the distance between neighboring lek territories was 14.8 ± 6.3 m. Territory size was 11.4 ± 4.4 nr. Territory size and distance between territories were among the lowest reported for Trochihnae hummingbirds. Lek attendance by territory owners fluctuated throughout the day. Activity slowly diminished after an initial period of activity after arrival at sunrise, but increased again between 0900 and 1 500 hrs. All males left their territories by 1830 hrs. Males sang at a similar rate (74.8 T. 14.5 songs/min) throughout the lekking season, but not throughout the day. There was no relationship between lek attendance and singing rate, two parameters that potentially affect mating success in lekking birds. Considerable interspecific variation occurs among lekking trochi lines, indicating that much remains to be investigated about lek behavior and structure in hummingbirds. Received 11 March 2011 Accepted W August 20J1. Lekking behavior has been described for at least 28 species of hummingbirds (Pizo and Silva 2001), and studies to date have revealed consid¬ erable interspecific variation in the structure and dynamics of hummingbird leks (Hdglund and Alatalo 1995, Ramjohn et al. 2003). For instance, the number of males at leks may vary from two for Rufous Sabrcwings {Campy lopterus rufus) (Skutch 1967) to >100 for Long-tailed Hermits (Phaethornis superciliosus) (Skutch 1964a), Lek¬ king males may be clustered, as with male Broad¬ tailed Hummingbirds ( Selasphorus platycercus) that remain ~7 m apart (Barash 1972), or form loose aggregations, as with male Swallow-tailed Hummingbirds (Eupetomena macroura) that arc 24-120 m apart (Pizo and Silva 2001 ). Activities at leks may continue throughout the day ( Phaethornis spp.; Ramjohn et al, 2003) or be limited to a short period such as dawn for Swallow-tailed Hummingbirds (Pizo and Silva 2001). Variation may reflect both phylogenetic and ecological constraints. However, Bleiweiss (1998) noted the repeated evolution of lekking behavior among hummingbirds, suggesting that behavior at leks is not limited by historical or phylogenetic constraints. Thus, to understand the relative roles of historical and ecological drivers in evolution of hummingbird leks, it is essential to expand our data base to include not only species representing different clades of the hummingbird phytogeny, but also to achieve a ' UNESP-Universidad Estadual Puulista, dc Zoologia, 13506-900 Kio Clam. SP. pizo@rc.unesp.br Departamcnto Brazil; e-mail; wider geographical sampling that encompasses contrasting ecological conditions. My objective was to examine the lek structure and lek behavior of Plovercrests ( Stephanoxis lalandi) in a subtropical area in southern Brazil. Stephanoxis is a monotypic genus of small hummingbirds (2.2-3.4 g. 8.5-9.0 cm), that occur in forests and semi-open areas from sea level to >2.000 m asl, ranging from eastern Paraguay to northeastern Argentina (Misiones). and southern ami southeastern Brazil (Schuchmann 1999), Males of the race .V. /. loddigesii that 1 studied have long iridescent blue crests with a conspic¬ uous black patch on the underparts, whereas females are iridescent green with gray below and with feathers of the head only slightly elongated. Hummingbirds in the genus Stephanoxis are known to display at leks (Sick 1997. Sigrist 2006). but details of their lekking behavior are unreported. Phylogenetically, the genus Stephu- noxis is placed within the Emerald clade (sensu McGuire et al. 2007), which includes genera that exhibit lekking behavior (e.g„ Amazilia and Campylopterus ) as well as apparently non-lekking species (e.g.. Thaluraniu and Chlorostilhon)- I also describe the song and aerial display of Plovercrests. their seasonal and daily patterns of activity at the lek. and the relationship between lek attendance (i.e., time spent in lek territories) and singing rate. METHODS Study Area.— The lek studied was at the edge of an early secondary forest (sensu Clark 1996) extending along the margin of a 4.5-km road connecting Ivoti and Lindolfo Collor (29 35' S, 106 Pizo . LEK BEHAVIOR OF THE PLOVERCREST 107 51 H' W; 130 m asl), two small towns in southern Brazil. The lek was in the transition area between ihe forest and an old field, where small trees and dvnhs ( Plovercrest" is among die lowest reported tor ^hi lines (Table 1 ). An additional reason tor the Putative preference for edge habitats, pertaining to lr°chilines, could be the importance ol an Equate light level for the proper exhibition of particular patches in the plumage of displaying birds, as demonstrated for other lekking birds (Endler 1996). Trochilines, contrasting with the usually drab colors of hermits, have iridescent patches in the plumage whose full brightness depends on the correct light level. One such iridescent patch in Plovercrests, the male crest, is conspicuously shown in lekking displays. The number of male Plovercrests at leks in my study (2-7) was within the range reported for othcr Trochilinae hummingbirds, which varies from two to 20 (Table 1). Territory size and distance between neighboring territories, howev¬ er. were among the lowest values reported for trochilines (Table 1 ). Lek size in hummingbirds is likely influenced by population density, being negatively correlated with distance between neiuhboring territories (Snow 1973). Territory size may also he influenced by vegetation structure. Stiles and Wolf (1979) noted the smallest territories of Long-tailed Hermits in Costa Rica were in very dense thickets, while the largest territories were in open forest under- story. The density of woody vegetation at the main Plovercrest lek was intermediate between a dense forest understory and an open field. Provided the observation by Stiles and Wolf ( 1 1)79) can be generalized, Plovercrest leks in forest interior should have territories closer to each other than reported in my study, Seasonal and daily activities at the Plovercrest lek in my study resembled those ol other hummingbirds. Activity at hummingbird leks is either limited to the breeding season (Snow 196S, 1974), with the exception of Swallow-tailed Hummingbirds that are found at leks throughout the year (Pizo and Silva 2001), or is greatly reduced during the non-breeding season (Stiles and Wolf 1979). Lek activities tor Plovercrests also coincide with the breeding period (Schuch- mann 1999, Belton 2003). The daily activity pattern of Plovercrests was similar to that described for White-bellied Emer¬ alds (Amazilia Candida ; Atwood et al. 1991). and Long-tailed Hermits (Stiles and W'olf 1979), where males left the lek after a period of high singing rate in the early morning. Singing rates again increased after this interval, but then gradually diminished during the hottest hours of the day. The daily activity pattern at hummingbird leks as hypothesized by Stiles and Wolf (1979) likely reflects the need of displaying males to leave the lek area for foraging. 110 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 by a> the lek. Parameters used in preparation of the sonogram: Win*. type-Hamming. n - 324 samples: Time gnd-Overlap 89.8%; Frequency grid-DFT, n = 2,048 samples. The aerial display of Plovercrests, in which a displaying bird hovers in front and slightly above a perched bird, is similar to the shuttle displays of North American hummingbirds (e.g.. Anna’s Hummingbird. Calypte anna ; Stiles 1982). The aerial display of Plovercrests appears to be similar to those of lekking hummingbirds in the subfamilies Trochilinae (e.g.. Swallow-tailed Hummingbirds, Pizo and Silva 2001) and Phaethornithinac (e.g.. Long-tailed Hermit Stiles and Wolf 1979). as well as those of son non-lekking species (e.g., Violet-capped Woo. nymph [Thai urania glaucopis |; M. A. Pizo. per obs.). This suggests the aerial display describe is deeply rooted in hummingbird phylogeny, ar not restricted to lekking activities. Interspecif differences in these displays do exist. Hermit for instance, often display their gape and thro patterns (Stiles and Wolf 1979), which apparent does not happen in Plovercrests. However, i properly evaluate these aerial displays and compiu jose of different species, recording the display with video cameras for more detailed analysis w be necessary. ■ I found no relationship between lek attendance by male Plovercrests and singing rate in conuast to White-bellied Emeralds (Atwood et al. 1991) These authors interpreted singing rates as reflecting dominance hierarchies among lekking males that might affect mating success. They noted, however, the relationship between lek attendance and singing rate was detected when the three leks studied were pooled for analysis, but became less clear when each lek was analyzed separately. Thus, variation existed among leks of a single species, and certainly exists in an interspecific comparison. No relationship between lek attendance and singing rate was detected for Plovercrests. but singing rates diltered among males. Relating differences to possible variation in mating success represents a challenge lor researchers because copulations are seldom witnessed even for the best studied lekking hummingbirds (Stiles and Wolf 1979). A better understanding of the relationship between lek behavior and mating success will require a combination of field effort to locate hummingbird nests near leks, and laboratory analysis to ascertain the paternity of nestlings. Pizo • LEK BEHAVIOR OF THE PLOVERCREST 111 ° Mean I I Mean±SE F MeaniSD Males FIG. 4. Box plots showing lek attendance (A) and singing rate (B) of seven male Plovercrests observed at a lek. Singing rates and territory attendance for each male were recorded during I - and 15-min periods, respective y. space our y ObOO to 1800 hrs. Sampling was spread over 20 days from August to December 2006. TABLE 1. Lek characteristics of Trochilinae hummingbirds. Ranges are presented whenever data for more than one individual territory and/or more than one lek were available. Species are in alphabetical order. _ Species Territorial males/lck Area of territories (nr) Distance between territories (m) Perch height (m) Source* ^tozilia amabilh 5 15-30 2.4-6 T 1C 6 9 A Candida A- Kacatl 3-7 2-4 <450 14- 32 15- 22 3-15 2-6 7 14 CmPyIopterits cumpeimis - ipennis up to 15 2-10 11. 12 2-4 10 C excellens 2-3 16 4 r- hemileucwvs 4 8 C hrgipennis 2-4 3 3 6 4 c nifus Petnmeiui mac r our a Wucharis eliciae K,ais guimeti l>na e°chroa cuvierii 2 15 3 4 4 133-266 24-120 <30 15 38 0.3-4.2 5- 13 6- 18 6-12 13 6 2 3 Maspharux platycercus U'Phanoxis lalandi 3 2-7 9-25 7 9-25 0.8-3.5 5 This study yaza pella 2-20 10.5 I ‘Sources' l r>n, .losxt ? skutch ( I9S8): 3. Skuich < 1964b): 4. Skuich ( 1967): 5. Barash (1972): 6. Skutch (1972): 7. Skuich ( 1981 ): 8. Hills and Broun "*6): 9, Atwood «al (1991);’ 10. Winker' e, al. (1992): II. Hayes el al. (1997); 12. Hayes el al. (2000); 13. Pizo and Silva (2001); 14. Hayes (2002). 112 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 ACKNOWLEDGMENTS I thank Luiz Pedreiru Gonzaga who prepared the sonogram, and K. L. Schuchmann, Johan Ingels, and an anonymous reviewer for critical comments on the manu¬ script. The author was supported by a research grant from the Brazilian Research Council (CNPq # 303559/2008-0). LITERATURE CITED Atwood, J. L.. V. L. Fir/, and J. E. Bamesbfrger. 1991. Temporal patterns of singing activity at leks of the White-bellied Emerald. Wilson Bulletin 103:373-386. Barash, D P 1972. Lek behavior in the Broad-tailed Hummingbird. Wilson Bulletin 84:202-203. Belton, W. 2003 Aves do Rio CJrande do Sul: distribuiyao e biologia. Editora Unisinos, Sao Leopoldo, Brazil. BleiweISS. R. 1998. Phytogeny, body mass, and genetic consequences of lek-maUng behavior in humming¬ birds. Molceulai Biology and Evolution 15:492-498. Brown. M. B. and A. B. Forsythe. 1974. Robust tests for the equality of variances. Journal of the American Statistical Association 69:264-267. Charif. R. A., C. W Clark, and K. M. Fristrup. 2007. Raven Pro 1.3 user's manual. Cornell Laboratory of Ornithology, Ithaca. New York, USA. Clark, D. B. 1996 Abolishing virginity. Journal of Tropical Ecology 12:735-739. Davis. T. A. W. 1958. The displays and nests of three forest hummingbirds of British Guiana. Ibis 100:31-39, Endler. J. A. 1996, Interacting effects of lek placement, display behavior, ambientlight. and color patterns in three neotropical forest-dwelling birds. American Naturalist 148:421. Hayes. F. E. 2002. Sabre rattling at the lek: morphological variation and its significance in the White-tailed Sabrewing (Campytopierus ensipennix). Pages 23-36 in Studies in Trinidad and Tobago ornithology honoring Richard FFrench (F. E. Hayes and S. A. Temple, Editors). Department of Life Sciences. University of the West Indies. Street Augustine, Occasional Papers Number 1 1. Hayes. F. F.., N. A. Trimm. B Sanasie, and R. p, FFRENCH. 2000. Breeding biology of the White-tailed Sabrewings at Tobago. West Indies. Journal of Field Ornithology 71:597-605. Hayes, F. E.. T. O. Garneit, M. V. Bernard, a. L. Bullard. D. R. Hardy. D.-A. D. Wilson. D. J. Wilson. V. L. JOSEPH, and D. K. St. Louis. 1997. Behavioral ecology of territorial male White-tailed Sabrewings ( Campytopierus ensipemis ): ev idence for lek polygyny Pitirre 10:27-28. Hilty, S. L. and W. C. Brown. 1986. A guide to the bir of Colombia. Princeton University Press, Princeto New Jersey. USA. Hoglund, J. and R. V. ALATALO. 1995. Leks. Princet. University Press. Princeton, New Jersey. USA. McGuire, J. A., c. C. Wrrr, D. l. Altshuler, and J. > Remsen Jr 2007. Phylogenetic systematic* an logLOgraphy of hummingbirds: Bayesian an maximum-likelihood analyses of partitioned data and selection of an appropriate partition strategy. Systematic Biology 56:837-856. Pl/O, M, A. AND W. R. Silva. 2001. The dawn lek of the Swallow-tailed Hummingbird. Wilson Bulletin 1 13:388-397. Ramjohn. C. L., F. B. Lucas. F. E. Hayes, S. T. Ballah. N. C. Johnson, and K. M. Garcia. 2003. Lek mating behavior of the Sooty -capped Hermit (Phaethmis augusti) in the Paria Peninsula of Venezuela. Journal of Field Ornithology 74:205-209. KUXTON, Cl. D. AND G. Beauchamp. 2008. Some suggestions about appropriate use of the Kmskal- Wallis test. Animal Behaviour 76:1083-1087. Schuchmann. K. L. 1999. Trochilidae Pages 468-680 u Handbook of the birds of the world. Volume 5. Barn- owls to hummingbirds (J. del Hoyo. A. Elliott, and J Sargatal. Editors). Lynx Edieions, Barcelona. Spain. Sick. II. 1997. Omitologia Brasileira. Editora Nova Fronteira. Rio de Janeiro, Brasil. StGRIST, f. 2006. Aves do Brasil, uma visao unislica. Author’s Edition, Sao Paulo. Brazil. Skutch, A. F. 1958. Life history of the Violet-headed Hummingbird. Wilson Bulletin 70:5-19. Skutch. A. E. 1964a. Life histories of hermit humming¬ birds. Auk 81:5-25. SKUTCH, A. F. 1964b. Life history of the Scaly-breasted Hummingbird. Condor 66:186-198. SKUTCH, A. F. 1967. Life histories of Central American highland birds. Publications of the Nuttall Ornitholog¬ ical Club Number 7. SKUTCH, A. F. 1972. Studies of tropical American birds. Publications of the Nuttall Ornithological Club Number 10, Skutch. A. F. 1981. New studies of tropical American birds. Publications of the Nuttall Ornithological Club Number 19. Snow. B. K. 1973. The behavior and ecology of hermit hummingbirds in the Kanaku Mountains. Guyana. Wilson Bulletin 85:163-177. Snow, B, K. 1974. Lek behaviour and breeding of Guy s Hermit Hummingbird Phaethomis guy. Ibis 116:278- 297. SNOW. D. W. 1968. The singing assemblies of Little Hermits. Living Bird 7:47-55. StatSoft. 1999. STATISTICA for Windows. StatSoft Inc . Tulsa. Oklahoma. USA. Stiles. F. G. 1 982, Aggressive and courtship displays of the male Anna’s Hummingbird. Condor 84:208-225 Stiles, F. G. and L. L. Wolf. 1979. Ecology and evolution of lek mating behavior in the Long-tailed Hermit Hummingbird. Ornithological Monographs 27:1-78. Winker. K. W.. M. a. Ramos. J. H. Rappole. and D W. Warner. 1992. A note on Campytopierus excelled in southern Veracruz, with a guide to sexing captured individuals. Journal of Field Ornithology 63:339-343. The Wilson Journal of Ornithology 1 24( 1 ): 1 13—1 18, 2012 NEST SURVIVAL, PHENOLOGY, AND NEST-SITE CHARACTERISTICS OF COMMON NIGHTHAWKS IN A NEW JERSEY PINE BARRENS GRASSLAND MICHAEL C. ALLEN1 AND KIMBERLY A. PETERS1 2 .ABSTRACT— We monitored Common Nighthawk ( Chordeilcs minor) nests in a managed grassland in the New Jersey Pine Barrens in 2009 and 2010. and assessed habitat selection by comparing vegetation characteristics at nests with random locations. We found relatively high nest survival with an estimated 79% chance of survival through incubation da. y survival rate = 0.987, n = 16 nests); predation was the most common cause of failure (n — 2). Movements o >ount up o 45 m from the original nest site) were frequent, which introduced uncertainty that prevented us bom estimating surviva through Hedging. Nest sites had significantly more open ground cover (e.g.. sand, lichen) than random sites, as we as ess shrub and grass cover, shallower Utter, and lower mean vegetation height. Received 9 May Jill . Aciepit - ugus The breeding biology and demography of the nightjars (Caprimulgidae) has been poorly studied worldwide relative to other groups (Straight and Cooper 2000, Holyoak 2001. Cink 2002, Brigham et ah 2011). The Common Nighthawk (C horde ties minor) is the most widely distributed and best- studied North American nightjar with a breeding range extending from Yukon, Canada to Panama (Brigham et al. 2011), but published data on reproductive rates, habitat preferences, and nest¬ ing phenology are lacking. Nesting usually occurs on the ground in a variety of open habitats including grasslands, gravel rooftops, and disturbed or open forests (Fowle 1946. Dexter 1952, Kantrud and Higgins 1992). The species is still common in many ;i,eas. but has exhibited a negative long-term population decline in the United States (Nebel etah 2010, Sauer et al. 2011), and has been assessed « ‘threatened’ in Canada (COSEWIC 2007). and of conservation concern in several U.S. states (e.g., New Jersey; NJENSP 2008). Only two published studies to our knowledge l|av'e quantitatively examined nest survival rates ot Common Nighthawks (Kantrud and Higgins '"2, Perkins and Vickery 2007), of which only 0ne used modern techniques involving daily survival rates (Perkins and Vickery 2007). Sim- llarY rnost data on reproduction by this species arc from urban rooftop nest sites (Bowles 1921: Sutton and Spencer 1949; Dexter 1952. 1956; Weller 1958; Dexter 1961; Armstrong 1965; Gramza 1967) with relatively few from natural ' New Jersey Audubon Society. Cape May Bird Obser- v;»°ry. 600 Route 47 North. Cape May Court House, NJ 982 1 0. USA. Corresponding author; e*mail; kpeters@massaudubon.org settings (Fowle 1946. Rust 1947, Kantrud and Higgins 1992. Perkins and Vickery 2007. Lohnes 2010). The only quantitative data on vegetation characteristics at nests in natural settings are from Kantrud and Higgins (1992) and Lohnes (2010). Data on reproductive rates, timing of nesting activities, and habitat preferences are important prerequisites to effect conservation actions for any species; these arc especially lacking tor the Common Nighthawk. We report data from Common Nighthawk nest monitoring in 2009 and 2010 in managed grasslands in the Pine Barrens of southern New Jersey. Our objectives were to: (.1) assess nest survival ami predation rates lor comparison with previous studies, (2) quantity nest-site character¬ istics and test whether they differed from those of surrounding available habitat, and (3) present location-specific information on clutch size, behavior of young, and phenology of nesting activities. METHODS Study Area.— Fieldwork was conducted on the - 3,000-ha Lakehurst section of Joint Base McGuire-Dix-Lakehurst in New Jersey, USA (40 q-> ' jq 74 ^ 22' W) within the boundaries ot the Pinelands National Reserve. Approximately 520 ha of the site are actively maintained as grasslands by mowing, burning, and mechanical shrub removal. All management activities at the site occur during winter or early spring which minimized disturbance to breeding grassland birds. Grasslands at the site occur in three main areas embedded within a landscape dominated by pitch pine (Pinus rigida ) and oak (Quercus spp.) forests. These are: (1) Test Sile-a 1 70-ha area surrounding a 3.5-km long runway that is rarely 113 THE WILSON JOURNAL OF ORNITHOLOGY • Voi 124. No. 1. March 2012 1 14 used, (2) Jump Circle-a 1 10-ha circular grassland used as an air-drop zone, and (3) Westfield-a 240-ha area encompassing two active airstrips. Fields are dominated mainly by warm-season grasses (e.g., Schizachyrium spp., Paniciun spp.) with significant amounts of bare ground (e.g., sand, lichens) and early-stage shrub encroachment (mainly P inus rigida and species in the family Ericaceae). Nest Survival and Predation Rates. — Nest searching occurred within 16 irregularly shaped plots totaling 257 ha (mean plot size = 16 ha, range = 9-32 ha). Plot boundaries were delineat¬ ed to provide an ample amount of searchable habitat (i.e., the maximum deemed feasible to search) within each of the three grassland areas of the base. Plot distribution was: six in Test Site (76 ha), four in Jump Circle (109 ha), and six in Westfield (71 ha). We searched two to three plots each weekday for ~2 hrs per plot beginning on 15 April (2009) and 28 April (2010). Searching involved one to three observers walking parallel transects and agitating vegetation with 2-m bamboo poles to flush nesting birds. Plot visits rotated so that each plot was searched at least once every 1-2 weeks. Geographic coordinates of located nests were obtained using a global positioning system, and two small pieces of pink flagging were placed on vegetation 2-3 m from the nest (i.e., creating a line with the nest at the center). Flagging was inconspicuous among surrounding vegetation, and intended to aid in relocation at close range; we do not believe that it drew the attention of predators. Nests were generally checked every 2-3 days until fledging, failure, or until young could no longer be located. Five check intervals for three nests exceeded 3 days due to logistical constraints, four intervals were 4 days and one was 5 days. Nest searching concluded on 15 July both years, although all active nests at that time were monitored until completion. Young nighthawks arc semi-precocial and te to move from the original nest site before fledgi with movements that appear to increase distance and frequency with age (Fowle 19< DeXt?rJ952; MCA, pers. obs.). We thorougl searched the area within -30 m of the nest (or i last Iocat,0n , which chjcks wcrc ohscrved in C,,'P,y "CStS- *» follow in most cases by at least one subsequent sear during the next nest-check. Typically" young T days post-hatch were easy to locate as they we invariably < 1 m from the original nest site, while older young could not always be found. Thus, wc calculated nest survival and predation raio. through hatch-date only (i.e.. success = hatching This is also the approach taken by Perkins am! Vickery (2007), who also worked in grassland habitat. Daily nest survival rates and confidence intervals were calculated using the logistic nest survival model within Program MARK (While and Burnham 1999. Dinsmore et al. 2002). Ths program was also used to evaluate whether or not daily nest survival rate varied over the course of the season. We used the likelihood ratio lest (Shaffer and Thompson 2007; alpha = 0.05) to evaluate model performance versus the null : i.e . constant survival or 'no effect’) model. Nesi failures were classified as either abandonment or predation based on timing and evidence at the nest. Daily predation rates were measured bv calculating the daily survival rate based on predation failures only, and subtracting this value from one. Clutch Size and Movements of Young.—1 Clutch size calculations were based only on active nests that were visited at least twice prior to hatching to avoid uncertainties associated with mobile young Observations made during checks after hatching were used to generalize pre-fledging movements of young from the original nest site. Nest -site Characteristics and Habitat Selection We measured maximum vegetation height (cm), after finding each nest, at which vegetation touched a meter-stick at five locations: at the nest and 0.5 n from it in each of the cardinal directions. Weal?1’ measured litter depth (i.e., unrooted, dead vegeta¬ tion) in 2010 at the same five locations. The mean value of the five measurements was used " subsequent calculations and analyses. We visual!) estimated the percent cover (±5%). alter noai - attempts were completed, of four vegetation caie gories within a 1 X 1-nr quadrat centered on tfc nest: (I) grasses (including rooted live and dc. grasses), (2) forbs, (3) shrubs (woody perennial- and (4) open (including sand. lichens, mosses. lilter). Only nests found prior to hatching were us* in vegetation calculations, as young found afio hatching may have wandered from the original " We also measured vegetation characteristics h 2010 at 80 randomly-generated points within '• 16 plots searched for nests to better assess habiM preferences. Points were generated in a geograPhK information system, and constrained to be at k1'1 Allen and Peters • COMMON NIGHTHAWK NESTING BIOLOGY 115 50 m apart to minimize spatial autocorrelation. Points were not used if they were on a road or other airfield infrastructure, and were substituted with the next point on the list so there were five points completed for each plot. All random vegetation measurements were performed between 1 5 and 23 June to coincide with the approximate midpoint of the grassland bird nesting season. Vegetation measurements were compared between years (2009 vs. 2010 nests), and between nests and random locations (2010 only) using nonparametric Wilcoxon rank sum tests (alpha = 0.05). Nesting Phenology. — We estimated the date of nest initiation (first egg laid) to assess nesting phenology for: (1) nests found during egg-laying i assuming 1 egg laid/day: Rust 1947), (2) nests at which the hatch date was known or could be estimated as the mid-point between two checks (assuming an 18-day incubation period: Brigham et al. 2011), and (3) nests found with young (estimated age based on photographs and field desenptions of known-age young; Brigham et al. 2011; MCA, unpubl. data). These methods provide adequate accuracy for estimating nest initiation (Nur et al. 2004). We acknowledge that our sample is biased as it excludes nests that were found and failed during incubation (i.e., egg- candling or floatation were not used to assign eggs to age classes), and therefore represents a dis¬ proportionate number of successful nests. We explicitly tested whether daily nest survival rates varied over the course of the season to address this concern. RESULTS Nest Survival and Predation Rates— We found -0 nests during the 2 years of the study: nine in 2009, and 11 in 2010. Four of the 20 nests (all from 2010) were excluded from nest survival analyses, including three found during the young s,age (i.e., post-hatching), and one found during the incubation stage that was inadvertently damaged by an observer during a check. We logged a total of 98 check intervals (median interval = 2 days) at the remaining 16 nests, yielding 224.5 exposure-days. 13)111660 of 16 nests included in nest survival analyses survived to hatching, while three tailed, two due to predation, and one was abandoned. All three nest failures occurred in 2010. The aban¬ doned nest apparently had infertile eggs as incubation was undertaken lor at least 27 days. The daily nest survival rate for the 16 nests was 0.987 (95% Cl = 0.960-0.996); thus there was -79% chance (95% Cl = 48-93%) of surviving an 18-day incubation period (calculated as [daily survival rate]1*). There was no evidence that daily nest survival rale varied over the course of the season (x = 1.1. df = L P = 0.29). The daily predation rate w-as 0.009, indicating there was a 15% chance of a nest being depredated during the 18-day incubation period (95% Cl — 4-47%). Eggs in 16 nests successfully hatched (including those found after hatching), and fledging was confirmed at only three. It is unknown whether young from the remaining 1 3 nests were predated or we lost track of them due to movements of young. The young at one nest were likely depredated by a northern pine snake { Pimophis mektnoleucus mel- anoleucus) that was observed ~ 1 m from 0-3 day old young that were not re-located. Clutch Size and Movements of Young.— C lutch size was two at 16 of the 17 nests found prior to hatching with the other nest containing a single egg. All three nests found after hatching contained two young. We recorded movements of young at 11 of 12 nests visited more than once. The one pair of young not observed to change locations was <5 days of age when last seen. Twenty-six movement events were observed in 42 post¬ hatching checks. Exact distances traveled were not uniformly recorded, but ranged during a single check interval (generally 2—3 days) from 0.15 to 6 m. Younger chicks tended to move less. Young that moved w'ere generally located close together (within 0.1 m) except for those close to fledging, which wrere at times — 1-2 m apart. One of four broods visited on day of hatching had not moved by the following nest check (2-3 days later), and three broods had moved only 15- 100 cm. The farthest straight-line distance recorded to the original nest site over the 14- day period one pair of young were monitored was —45 m. In contrast, a pair of 1 1-day old young at another nest moved only 0.5 m from the original nest site. Nest-site Characteristics and Habitat Selection— Nest sites in both 2009 and 2010 were dominated by open ground (mean ± SD cover. 58 ± 20%, range = 5-80%. n = 17), followed by grass ( 18 ± 11%, range = 5-35%). shrubs (9 ± 13%, range = 0- 40%). and forbs (9 ± 15%, range = 0-45%). Mean vegetation height at nest sites wus 11 ± 7 cm (range = 0-50 cm). Vegetation characteristics at nests were 1 16 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 Open Grass Shrub O O Forb FIG. I. Percent vegetative cover at random locations (white boxplots, n = 80) and Common Nighthawk nests (gray boxplots, n = 8.) on Joint Base McGuire- Dix-Lakehurst, New Jersey. USA, in 2010. Open ground included bare sand, lichens, and matted dead vegetation (litter). Boxplots show median, interquartile range, and range of data. Open circles show points >1.5 interquartile ranges above the median. similar in 2009 and 2010 (Wilcoxon rank sum tests, IV = 24-43. P > 0.26), but 2010 nests diftered from vegetation at random locations (Fig. 1, Table I). Nest sites in 2010 had more open ground, shorter vegetation, shallower litter, and less grass and shrub cover than random areas' Nesting Phenology.— We estimated initiation dates for 16 of the 20 nests. Four were found during egg-laying, two were estimated based on a known hatch date, seven from an estimated hatch date, and three from the estimated age of young. The median nest initiation date was 31 May with an interquartile range of 25 May to 11 June, and a range of 18 May to 28 June (Fig. 2). One nest for which we could not accurately estimate initiation date was estab¬ lished prior to 17 May, as it was discovered with a two-egg clutch on this date. 120 1 40 160 180 Julian date FIG. 2. Estimated initiation dates (first egg Iaidi for Common Nighthawk nests monitored on Joint Base McGuire-Dix-Lakehurst, New Jersey. USA, in 20n° 2010. Bins are in increments of 5 days, beginning on I' May (Julian day 135). Boxplot above shows median, interquartile range, and range of dates. Asterisks indican the average start and end dates for nest searches in 200' and 2010. DISCUSSION The lack of data on breeding biology toi nightjars is a significant obstacle to effective conservation planning. This is especially relevant in North America as several species appear to I experiencing long-term population declines (Wilson 2008, Nebel et al. 2010. Sauer el al 201 1 ). We found nest survival through incubati"’ to be 79% (0.987 daily survival) which b considerably higher than reported by Perkins anJ Vickery (2007) in Florida (28%, 0.932 daily survival, n = 14 nests). Sample sizes for both studies were somewhat lower than those general recommended for daily survival rate estimation * SE) “ Comm°" nest sites (2009-2010) *1 - Wilc„r„ raTsUm fLr^n1 P'°‘S “ J°inl Bs“ McGttinr-Dix-Lakehurst. New Jersey. USA. - afe displayed between the columns: ns, P > 0.05; *, 0.05 > P > 0.01 : **, 0.01 >P>(>' _Yeg- height (cm) % Open % Grass % Forb % Shrub Litter depth (cm) Nests 2009 (« = 9) 8.6 ± 3.3 53.9 ± 7.9 2LO ± 3.7 7.8 ± 4.3 6-7 ± 3.4 Nests 2010 (n = 8) ns ns ns ns ns 9.7 ± 1.8 63.1 ± 4.8 14.4 ± 3.5 10.6 ± 7.0 12.5 ± 5.6 0.3 ± 0.1 ** ** * ns Random 20l0^jjy_ 21.5 1 14 31.8-24 40.4 - 3-0 3.1 i 1° 29.1 t 3.0 1.2 1 0-1 Allen and Peters • COMMON NIGHTHAWK NESTING BIOLOGY 117 (i.e„ >20 nests; Hensler and Nichols 1981), which is reflected by the wide confidence intervals around these estimates (48-93% incubation .survival for nur study, and 11-68% in Perkins and Vickery 2007). We found apparent nest survival (i.e., % successful nests) of 81% (13 of 16 nests) in comparison to 93% in grasslands of the northern Great Plains (13 of 14 nests [excluding 1 human- induced failure): Kantrud and Higgins 1992). and 43% in Florida dry prairie (6 of 14; Perkins and Vickery 2007). Neither Kantrud and Higgins 1 1992) nor Perkins and Vickery (2007) reported nest abandonment and both concluded that preda¬ tion was the main cause of nest failure. Published studies of Common Nighthawks have considered only nest survival through hatching. A complete picture of nest survival requires data for the period of young development. Including this ■uage would necessarily result in lower estimates of overall success, If we assumed, for example, the same daily survival rate for the period of young development as we found for incubation, the expected probability of survival to fledging at our site would be 62% (assuming 18 days front hatching until Hedging; Brigham et al. 201 1 ). The question of whether survival differs substantially between eggs and pre-fledged young in Ibis species will likely require telemetry data due to ihe uncertainties associated with monitoring semi- precocial young (e.g., Fowle 1946, Rust 1947. Perkins and Vickery 2007). Clutch size of Common Nighthawks is similar across a broad geographic area with two-egg clutches dominant in Idaho (24 of 27 clutches; Rust 1947), the northern Great Plains (18 of 21; Kantrud and Higgins 1992). New Jersey ( 16 of 17; ihis study), and Florida (13 of 14; Perkins and Vickery 2007). All other clutches consisted of one et'g- Perkins and Vickery (2007) argued that some one-egg clutches could be the result of partial depredation, and both Rust ( 1947 ) and Sutton and Spencer (1949) observed eggs rolling from nests when ihe female flushed, which could be another source of clutch reduction. We observed one nest la 2-egg clutch) at which the second egg was not incubated, but was found •— 1 m distant. The movements of young we recorded were similar to previous reports for both rooftop and non-rooftop sites (Fowle 1946. Rust 1947, Dexter 1952). Fowle (1946) reported single-day move- ments as far as 15-27 m in a burned clear-cut on Vancouver Island. British Columbia, compared to 'he maximum of 6 m in 2 days in our study. It is possible that Fowle’s handling of young for weighing (not done in our study) contributed to the farther single-day movements he observed. Few data exist on the habitat preferences of Common Nighthawks. However, our finding that open ground was preferred is not surprising based on several qualitative accounts (e.g., Fowle 1946, Rust 1947. Brigham et al. 2011) and two quantitative studies (Kantrud and Higgins 1992, Lohnes 2010). Typical nesi sites in our study were in patches of ‘open’ ground (e.g., sand, lichens, litter), between warm-season grasses or erica- ceous shrubs that often provided partial shade. Lohnes (2010) compared nest sites with random areas in the Konza Prairie in Kansas and also found a preference for open ground. Kantrud and Higgins (1992) noted that over half of 21 Common Nighthawk nests in the northern Great Plains had 'no vegetation' and they report an average vegetation height ot 6 cm, considerably lower than that of other ground-nesting birds in the area (mean = 33 cm). We found mean vegetation height at nest sites to be about half that of random areas (10 vs. 22 cm), a discrepancy at least partly driven by the higher number of zero height values at nest sites in open areas. The pattern of nest- initiation dates observed in our study appeared to be unimodal (Fig. 2), although it is possible that increased sample sizes would reveal a different pattern. Some nests also may have been initiated before or after nest searches (15-28 Apr to 15 Jul). This is not likely to be a significant proportion of nests, however, as nighthawks do not typically arrive on the study site until early to mid-May (MCA, pers. obs.) and fall migration in this species begins in mid-August (Walsh et al. 1999). The median initiation date we observed (31 May) was earlier than observed in the northern Great Plains (24 Jun, n = 8 nests, Kantrud and Higgins 1992) and northern Idaho (30 Jun, n = 27; Rust 1947), hut our range (18 May-28 Jun) was within the range observed in these studies (7 May- 15 Jul). ACKNOWLEDGMENTS This study w as funded by the Department of Defense Legacy Resource Management Program and ihe U.S. Navy Agricultural Outlease Program. Field work was performed by Mike Allen. Ron Hutchison. Tamarra Mart/, Kim Peters, Ben Sandstrom. Katie Schill. I.ena Usyk, and Rachel Villani. We thank John Joyce of Joint Base McGuire- Dix-Lakehurst for helpful logistical support. 1 18 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 LITERATURE CITED Armstrong, J. T. 1965. Breeding home range in the nighthawk and other birds: its evolutionary and ecological significance. Ecology 46:619-629. Bowles. J. H. 1921. Nesting habits of the nighthawk at Tacoma. Wash. Auk 38:203-217. Brigham. R. M.. J. No. R. G. Poulin, and S. D. Grindal. 2011. Common Nighthawk ( Chordeiles minor). The birds of North America. Number 213. Cink. C L. 2002. Eastern Whip-poor-will I Caprirnulftus vociferus). The birds of North America. Number 620. Committee on the Status of Endangered Wildlife in Canada (COSEWIC). 2007. COSEWIC assessment and status report on the Common Nighthawk Chor¬ deiles minor in Canada. Committee on the Status of Endangered Wildlife in Canada, Ottawa. Ontario, Canada. Dexter, R W. 1952. Banding and nesting studies of the Eastern Nighthawk. Bird Banding 23:109-1 14. Dexter. R. W. 1956. Further banding and nesting studies ol Ihe Eastern Nighthawk. Bird Banding 27:9-16. Dexter. R. W. 1961. Further studies on nesting of the Common Nighthawk, Bird Banding 32:79-85, Dinsmore. S, J.. G. C. White, and F. L. Knopf. 2002. Advanced techniques for modeling avian nest survival Ecology 83:3476-3488. Fowix, C. D. 1946. Notes on the development of the Common Nighthawk. Auk 63:159-162. Gramza. A. F. 1967. Responses of brooding nighthawks to a disturbance stimulus. Auk 84:72-86. Hensler, G. L. and J. d. Nichols. 1981. The Mayfield Method of estimating nesting success: a model, estimators and simulation results. Wilson Bulletin 93:42-53. Holyoak. D. T. 2001. Nightjars and their allies. The Capri mulgiformes. Oxford University Press, Oxford. United Kingdom. KantRUD, H. A. and K. F. Higgins. 1992. Nest and nest site characteristics of some ground-nesting, non- passertne birds of northern grasslands. Prairie Natu¬ ralist 24:67-84. Lohnes, R. G. 2010. Nest site selection and nest thermal properties of Common Nighthawks on the tallgrass prairie of Kansas. Thesis. Cornell University, Ithaca New York. USA. Nebel, S., A. Mills, J. D. McCracken, and P. D. Taylor. 2010. Declines of aerial insectivores in North America follow a geographic gradient. Avian Grow vation and Ecology 5:1-14. New Jersey Endangered and Nongame Species Pwoc" (NJENSP). 2008. Special concern-species statnslicini 27 October 2008. New Jersey Division of Fl4i air, Wildlife, Trenton, USA. http://www.5LUe nj.ucdrr fgw/ensp/pdf/spclspp.pdf Nur, N„ A. L. Holmes, and G. R. Geupel. 2004 Use survival time analysis to analyze nesting success r birds: an example using Loggerhead Shrikes. Condo: 106:457-471. Pi-.kkin.s, D. w. and P. D. Vickery. 2007. Nest success "i grassland birds in Florida dry prairie. Southeaster Naturalist 6:283-292. Rust, H. J. 1947. Migration and nesting of Common Nighthawks in northern Idaho. Condor 49: 177- 188 Sauer, J. R„ J. E. Hines, J. E. Fallon. K. L. Pardibl D. J. ZiOLKOwsKi Jr., and W. A. Link. 2011. Tic North American Breeding Bird Survey, results ana analysis 1966-2009. Version 3.23.2011. USGS, Pa¬ tuxent Wildlife Research Center. Laurel. Mankind USA. http://www. mhr-pwrc.usgs.gov/bbs/ Shaffer, t. L. and F. R. Thompson III. 2007. Making meaningful estimates of nest survival with mode! basetl methods. Studies in Avian Biology 34.84-93. Straight, C. A. and R. J. Cooper. 2000. Chuck-will ' widow (Caprimulgits carolinensis). The birds ot North America. Number 499. Sutton, G. M. and H. H. Spencer. 1949. Observations at a nighthawk's nest. Bird-Banding 20:141-149. Walsh, j.. V. Elia. R. Kane, and T. Haluweu. 1999- Birds of New Jersey, New Jersey Audubon Societ> . Bemardsvillc, USA. Weller, M. W. 1958. Observations on Ibe uicuhauon behavior of a Common Nighthawk. Auk 75:48-39 White. G. c. AND K. P. BURNHAM. 1999. Program MARK survival estimation from populations of royLJ animals. Bird Study 46( Supplement): 1 20- 1.^ Wilson, M. D. 2008. The Nightjar Survey Network program construction and 2007 Southeastern nightjar survey results. Technical Report Series CCBTR- v 001. Center for Conservation Biology. Williamsburg Virginia, USA. The Wilson Journal of Ornithology 124(1 ):1 19-126. 2012 REPRODUCTIVE LIFE HISTORY TRAITS OF THE YELLOWISH PIPIT (. ANTHUS LUTESCENS) MAIKON S. FREITAS1 AND MERCIVAL R. FRANCISCO12 ABSTRACT.— We describe reproductive traits of the Yellowish Pipit (Anthus lulescens) in the Slate of Sao Paulo. Brazil. We found 32 active nests during three breeding seasons (2008-2010). Domed nests were built exclusively on the ground where the grass was sufficiently tall to conceal them. Clutch initiation across years occurred from July to October and average ± SD clutch size was 3.05 ± 0.4 eggs or young. Yellowish Pipits were predominantly single-brooded. Eggs were pale white with brown spots and blotches that could be more concentrated at the larger end or homogeneously distributed over the entire surface. Eggs were 18.2 1 0.8 mm in length. 13.7 - 0.3 mm in width, and weighed 1.7 _0.1_g. Incubation and nestling periods lasted 1 3.03 ± 0.2 and 1 4.5 ~ 1 .0 days, respectively Mean lime incubating/hr was 38 ± 71 min, and incubation recesses averaged 9.4 t 4 nun. Young were provisioned on average 13.3 £ 7.9 Utne. , > males and females. Estimated overall nesting success using a null model of constant nest survival rates was X e ( - c 36-97*5). Model selection analyses indicated survival was negatively correlated to nest age and time within the breeding reason. Comparisons of Yellowish Pipit life history traits with northern temperate congeners provided support for the premises that clutch sizes are smaller and young development is slower in the tropics. The hypothesis that annual fecundity ;an be similar across latitudes due to a negative correlation between clutch size and number of renesting attempts was not supported. Our data contradicted the commonly claimed, but poorly tested hypothesis, that smaller clutch sizes in the tropics can he explained by a longer breeding season that permit more opportunities to rencst within the same breeding reason. Received 14 February 2 Oil. Accepted 15 August 201 1. Several studies have addressed avian life history adaptations of northern temperate versus iropical and southern temperate habitats. Broad latitudinal patterns of reproductive traits have been proposed, i.e.. Southern Hemisphere species have smaller clutch siz.es (Moreau 1944, Lack 1947. Skutch 1949, Murray 1985). lay more clutches per year (Lack and Moreau 1965, Ricklels 1969). and have longer incubation and nestling periods (Skutch 1949). The occurrence of latitudinal differences in clutch size has been tested (Yoni-Tov et al. 1994. Young 1994. Geffen and Yom-Tov 2000, Martin et al. 2000. Ghalam- bor and Martin 2001). but the claim thal incubation and nestling periods, as well as number °l renesting attempts, differ between Northern and Southern hemispheres is still not accepted (Geffen and Yom-Tov 2000). Only a few studies have compared species paired by phylogeny and ecology, and attempted to isolate the latitudinal effect from the phylogenetic influence (Yom- Tov et al. 1994. Marlin et al. 2000, Ghalambor and Martin 2001, Martin 2002). Passerines of lfle genus Anthus (Motacillidae) are globally distributed, and occur on every continent ex- Cept Antarctica (Ridgcly and Tudor 1994, Tyler -1*94). This makes them well suited for studying 1-niversidade Federal de Sao Carlos, Campus de Surocaba, Rod. Joao Lome dos Santos, km 110, Some-aba. Sp 18052-780. Brazil. Corresponding author; e-mail: mercival@utscar.br breeding trail diversification across latitudes. However, many motacillids have not been studied in detail, especially those in South America (Tyler 2004). We present the first comprehensive description of the reproductive life history traits ot the South American Yellowish Pipit (A. lutescens). Our objectives were to: (1) provide information on phenology and duration ot breeding season, clutch size, length ot incubation and nestling periods, nesting success, renesting attempts, and parental care for a Sao Paulo State population, southeast Brazil; and (2) compare these life history traits with data from the literature lor a set of northern temperate congeners. METHODS Study Area. — Observations of Yellowish Pipits were conducted along the Sorocaba River and in an adjacent urban park (16 ha), including an artificial’ lake of 2.8 ha. The study area is in the suburbs of the city of Sorocaba, State of Sao Paulo in southeast Brazil (23 28' 99" S, 47 26' 17" W). The area is characterized by the presence of large patches of exotic grasses (mainly Zoysia japonica and Cynodon daetylon). Trees and bushes are widely spaced and the grass is kept relatively short by employees of Sorocaba city prefecture, which provided suitable breeding habitat for Yellowish Pipits. The climate is tropical with two well-marked seasons: a humid, hot season 119 120 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 from October through March (average rainfall = 919 mm. temperatures varied from 15.7 to 32.4 °C) and a dry. cold season from April through September (average rainfall = 294 mm, temper¬ atures varied from 11.4 to 30.6 C). Study Species.— The Yellowish Pipit is a monomorphie passerine widely distributed in South America from western Panama south to Uruguay and Argentina (south to La Pampa and Buenos Aires). It inhabits grasslands, open C'er- rado. pastures, and cultivated lands often near water or marshes (Ridgely and Tudor 1994, Sick 1997, Tyler 2004). The Yellowish Pipit is a common species, but data on its breeding biology are scattered and many aspects are poorly docu¬ mented. Data on incubation and nestling periods, breeding phenology, and parental care are present¬ ed here for the first time. Field Procedures. -Wc conducted systematic nest searches from July to January during three breeding seasons (2008, 2009, and 2010). We searched for nests at least three times per week by covering 1 km of each margin of the Sorocaba River and Lhe entire adjacent park. Nests were located by following adults in their territories when they were carrying nest material or feeding young, and were checked every 1-3 days. We used metal calipers accurate to ±0.01 mm to measure nests and eggs, and a spring scale accurate to ± 0. 1 g to weigh eggs. The incubation period was from the first day of incubation to the day before hatching, and nestling period was from hatching day to the day before Hedging. Observations were performed daily during the laying stage, and we could detect females that began incubation before and after the set of eggs was complete. We also checked if eggs were warm to detect Lite beginning of incubation. We did not touch or handle young to avc shortening the nestling period (Skutch 194‘ Clutch initiation dates were obtained from ne: found in die construction stage (i.e.. we observ the first egg in die nests) (n = 15), and by bac dating for nests for which hatching or fledgi dates were known, based on mean incubation u nestling periods (// = 12), We estimated lhe frequency at which adu >i '° WM thc Pulton = 15). Nests were domed with a small side entrance, often invisible from above. Nests were exclusively on the ground, where the grass was sufficiently tall to conceal them, usually in slight depressions in the soil under dense vegetation Fourteen nests were on flat terrain, while 18 were in crevices in the banks of the Sorocaba River. 2-5 m from water. Nest material consisted ot dry grasses, and the nest chamber was lined with finer grass leaves and grass stems (Fig. I). Nest measurements varied (Table 1 ) and one nest had a tubular entrance ~13 cm in length. Males and females shared nest building activities with both carrying and placing nest materials, often one after the other at similar rates (MSF, pers. obs.). Adults brought nest materials on average 12.6 ± 8.4 times/hr (range = 4-24) during 7 hrs of focal observation (n = 6 nests). Time gaps between visits were 3.6 ± 4.7 min in length (range = 0.1- 25, n = 88 observations). Two nests found at the beginning of construction took 3 days to complete. The earliest clutch initiation dates (laying of the first egg) varied among years: 20 August 2008, 6 September 2009. and 28 July 2010 (Fig. 2). The latest clutch initiation was on 19 October 2008, and the latest nesting activity (the last young observed in a nest) was on 4 November 2008. The average breeding season (pooling the 3 yrs together) spanned almost 5 months, but nests found in July and November were exceptions (only I nest each). Laying initiations were highly concentrated in August and September, and the average number of active nests was concentrated in August through October (Fig. 3). However, clutch initiations were clearly concentrated in only 1 month (Fig. 2) when each year was analyzed separately. The basic egg color was pale white and markings varied remarkably, even within a single nest. Some eggs were predominantly white with pale to intense brown spots and blotches moie concentrated at the larger end. while others were heavily spotted over the entire surtace (Fig. 1). Eggs measured 18.2 ± 0.82 mm in length (range = 17.1-19.7) and 13.7 ± 0.3 mm in width (range = 13,3-14.2), and weighed 1.7 ± 0.12 g (range = 1 5-2.0) (n = 16). Clutch sizes were two ( n = 1), three (n = 18). or four (n = 2) eggs or young (3.05 ± 0.4), and eggs were invariably laid on consecutive days (n = 15 nests). Incubation oi 10 observed nests, started on the third day after onset of laving, even when clutch sizes were two ( n — 1 nest) or four eggs (rt = 1 nest). One egg took 14 days to hatch, but the incubation period was 13 days (29 eggs from 10 nests) (13,03 ± 0.2). We believe that only females incubated for two reasons: (1) we did not observe adults taking turns TABLE 1. Measurements (mm) of Yellowish Pipit nests length (OL). inside high (IH), inside diameter (ID), entrance („ = 1 1 ): outside high (OH), outside width (OW), outside width (EW), and entrance high (EH). ow OL IH ID EW EH Mean ± SD Range 88 ± 8.3 71-104 114.4 ± 11.7 130.25 ± 15.6 93-127.5 104.5-157.3 62 ± 8 50.1-74.7 82.4 ± 16.7 55.8-108.5 56 ± 9.9 37.7-68 46.5 ± 98.8 30-58.6 71 122 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. 1. March 2012 w c a) -O E 12 10 8 6 4 2 0 □ 2008 ■ 2009 250 Months -• Number of clutch initiations of Yellowish Pipits across different months through three breeding seasons (2008. 2009, and 2010), and monthly rainfall. to incubate; and (2) the non-incubating individual often remained near the nest singing vigorously, suggesting it was the male. Males did not feed females in the nests in 3 1 hrs of focal observations at nine different nests, bui usually escorted them during incubation recesses. Females spent from 23.2 to 55.8 min incubating the eggs per hr (38 ± 7.1 min). They left the nests 1-3 times per hr (2.2 ± 0.6). and incubation recesses were 3.8 to 26.3 min in length (9.4 ± 4, n = 53). When incubating females were Hushed from nests, they performed a ‘broken-wing’ display, vocalizing and dragging their wings along the ground in an attempt to distract the observers. Hatching was synchronous (nests were checked from 0900 to 1000 hrs) in 12 of 16 nests. Nestlings had yellowish skin and were covered with gray down at hatching. The bill and swollen flanges were yellow and the mouth lining was blight orange. Adults gathered a large number of small 25 20 15 10% (Craig and Enderson 2004:fig. 46). Similar !' rates varied between 64 and 86% in Washingt1" TABLE 2. Occupancy rates of Peregrine Falcons at nest-sites in Colorado. Montana, and Wyoming. 2005-2009. Nest-sites studied Year CO MT WY 2005 2006 2007 2008 2009 Means 29 36 27 27 44 64 80 86 93 1 12 64 61 54 29 46 Sites with pairs CO MT WY 27 52 64 34 67 61 23 68 51 23 74 29 41 84 41 co 93 94 85 85 93 90 Occupancy ('*) MT WY 81 84 79 80 75 80 100 100 94 100 89 97 Enderson et al. • PEREGRINE FALCON NESTING PERFORMANCE 131 TABLE 3. Nest success and reproductive performance of Peregrine Falcons in Wyoming, 2005-2009. Colorado. Montana, and Yea Number of pairs" Successful pairsh Success rale (%) Number of young Young per pail" CO MT WY CO MT WY CO MT WY CO MT WY CO MT WY 2005 24 49 64 16 40 45 67 82 70 36 94 99 1.5 1.9 1.5 2006 32 66 61 24 58 44 75 88 72 53 147 101 1.7 2.2 1.7 2007 17 68 51 12 51 36 71 75 71 28 108 75 1.6 1.6 1.5 2008 11 67 29 8 54 19 73 81 66 13 125 45 1.2 1.9 1.6 2009 28 79 41 26 69 28 93 87 68 61 176 58 2.2 2.2 1.4 Totals 112 329 246 86 272 172 191 650 378 Means*3 77 83 70 1.7 2.0 1.5 ! Pairs for which reproductive outcome was known. ’ Pairs producing at least one young 28 days of age or older. c Based on all pairs of known outcome, successful or not. , 0 Mean success rale based on the total number of successful pairs. 2005-2009, divided by the total number of pairs. Slate. 1990-2001. with differences between con¬ secutive years as great as 12% (Hayes and Buchanan 2002:table 4). The USFWS national monitoring results for 2003 show a rate of 87% for 90 territories in Colorado, Montana. Utah, and Wyoming (Green et al. 2006), and 81% for 91 territories in the same slates in the draft report for 2006 (M, G. Green, unpubl. data). Occupancy rate may not be a sensitive indicator of population change because of considerable between-year variation. The vagaries of adverse weather may explain some variation in nest success. Wet weather, in -009. probably caused nest failure at three sites in the Bighorn Mountains of Wyoming, a phenom¬ enon reported elsewhere (Olsen and Olsen 1989). The nest-sites we studied were distributed across a •arge region and adverse weather in any year would not likely affect nesting in all states, or even in parts of states, to the same extent. Annual nest success in Washington, in compar¬ ison. varied widely between 40 and 80% (mean — 61%,/? = 460) in 1990-2001 (Hayes and Buchanan -002). Success may be inflated in that study because young of all ages were recorded. Annual nest success in Idaho. 2005-2009, w as between 52 and 83% ( mean = 71%.//= 1 29 ) ( Moulton 2009 ); °nly young —33 days of age or older were recorded. The success rate for 70 pairs in Colorado. Montana, Utah, and Wyoming in the 2003 USFWS monitoring survey was 74% (Green et al. 2006). and the draft report for 2006 gave a success rate of 70% for 70 pairs (M. G. Green, unpubl. data). Wide variation in reproduction rate among years has been reported. Reproduction rate in Colorado ranged from 1.4 to 2.1 (mean = 1.7) young/pair (40-70 nesting attempts/year) during 1995-2001 (Craig and Enderson 2004). Repro¬ duction rate for 10 pairs in 2004 averaged 2.1 young/attempt (Enderson 2005). Reproduction rate in Washington in 1990-2001 varied from 1.0 to 2.2 young/pair, but may have been biased upwards because voting of all ages were reported (Hayes and Buchanan 2002). In Washington, 449 young were counted in those years in 679 nesting attempts (mean = 1.5). Reproduction rates during annual counts in the same period in Idaho averaged between 1.0 and 2.5 (6 to 15 nesting attempts/year) (Moulton 2009). The 2005-2009 average in Idaho was 1.6 young/pair for 129 nesting attempts. The 2003 USFWS national survey found a reproductive rate of 1.5 young/ pair in Colorado, Montana, Wyoming, and Utah (Green et al. 2006) and 1.4 in 2006 (M. G. Green, unpubl. data). The 2000 Canadian na¬ tional peregrine survey reported a mean ol 2.5 young per pair (n — 23) in Alberta south of 58 N latitude, a population where some pairs were adjacent to those in Montana (Rowell et al. 2003). We conclude that annual reproduction in the range of 1.4 to 2.0 young/pair on territory was usual. Wide year-to-year fluctuations were common, perhaps caused partly by adverse weather. Reproduction measured in our study seems robust, but factors such as adult mortality, age at first reproduction, and immigration, all usually unknown, combine to affect population change. Craig et al. (2004) modeled these components based on values from Colorado during 1989- 2001. They predicted an average reproduction of — 1.7 young/pair would yield a population growth 132 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 rate of 3 to 8% per year, the range resulting from the actual values accepted for the other compo¬ nents. That prediction coincides with our discov¬ ery of new pairs each year at cliffs where none had been seen in earlier searches. ACKNOWLEDGMENTS This work was supported in part by funds from the U.S. Fish and Wildlife Service administered by the Montana Fish, Wildlife and Parks Department, the Wyoming Game and Fish Department, and the Colorado Division of Wildlife, and partly by funds arising within these agencies. K. C Uaynam and Adam Shteading helped with fieldwork in Montana. Observers in Wyoming included Terry McFaneaney, D. R. Mutch. S. M. Patla, D. W Smith, and L, A. VanFlcct. Michelle Cowardin, David Klute. K. M. Potter, and Michael Reid participated in Colorado. We arc grateful for their help. We thank Colleen Moulton for permission to use recent information from Idaho. Reviews by M. R. Fuller and C. M. White improved this paper. LITERATURE CITED Burnham. W. a., W. Heinrich, C. Sandport, E. Levine. D. O brein, and D. Konkel. 1988. Recovery effort for the Peregrine Falcon in the Rocky Mountains. Pages 565-574 in Peregrine Falcon populations, their management and recovery (T. J. Cade, J. H. Enderson, C. G. Thelander. and C. M. White, Editors). The Peregrine Fund Inc.. Boise, Idaho, USA. Cade. T J. and W. A. Burnham (Editors). 2003. Return of the peregrine. The Peregrine Fund. Boise. Idaho, USA. Craig. G. R. .and J. H, Enderson. 2004. Peregrine Falcon biology and management in Colorado. 1973-2001. Technical Publication Number 43. Colorado Division of Wildlife. Fort Collins, USA. Craig, G R., G, C. White, and J. H. Enderson. 2004. Survival, recruitment, and rate of population change of Peregrine Falcons in Colorado. 1963-2004. Journal of Wildlife Management 62:1032-1038. Enderson, J. H. 2005. Changes in site occupancy am) nesting performance of Peregrine Falcons in ColornF Journal of Raptor Research 39:166-168. Enderson, J. H. W. Heinrich, L. Kin-, andC. M. Wttnt 1995. Population changes in North Amcncin pere¬ grines. Transactions of the North American Wildlife and Natural Resources Conference 60:142-161 Green. M. G„ T. Swem. M. Morin, R. Mesta. M Kiel K. Holar, R. Hazelwood, P Delphev, R. Cwril and M. Amaral. 2006. Monitoring results foi breeding American Peregrine Falcons (Falco pertpi- nus anatom), 2003. Biological Technical PuWiatirm FWS/BTP-R 1005-2006. USD1, Fish and Wildlife Service, District of Columbia. Washington, USA Hayes. G. E. and J. B. Buchanan. 2002. Washington State status report for the Peregrine Falcon. Wavlungior Department of Fish and Wildlife. Olympia. USA. Moulton, C. (Editor). 2009. Idaho Peregrine Falcon survey anil nest monitoring 2009 Annual Report Mato Department of Fish and Game. Boise, USA. OLSEN. P. D. and J. Olsen. 1989. Breeding of th; Peregrine Falcon (Falco peregrinus). III. Weather, nest quality and breeding success. Emu 89:6-14 Rowell., p„ G. L. Holroyd. and U, Banasch (Editors 2003. The 2000 Canadian Peregrine Falcon sums Journal of Raptor Research 37:98-116. Sti.enhof. K. and I. Newton. 2007. Assessing nesting success and productivity. Pages 181-192 in Raptor research and management techniques (D. M. Bird and K. L. Bildstein. Editors). Hancock House Publishers. Surrey, British Columbia, Canada. U.S. Department of the: Interior (USD!). 1999. Endan¬ gered and threatened wildlife and plants: final rule to remove the American Peregrine Falcon from tilt Federal list of endangered and threatened wildlife Federal Register 64: 46542-46558. U.S. Department of the Interior (USD1). 2003 Moni¬ toring plan for the American Peregrine Falcon. 2 species recovered under the Endangered Species Ad. USD1, Fish and Wildlife Service. Divisions ot Endangered Species, Migratory Birds, and Stale Programs. Pacific Region. Portland. Oregon. USA. The Wilson Journal of Ornithology 1 24( 1 ): 1 33-1 38, 20 1 2 reproductive success of the creamy-bellied thrush in a SOUTHERN TEMPERATE ZONE ANDREA ASTIE1-2 AND NATALIA LUCHESI1 ABSTRACT— We describe ihe breeding biology and reproductive success of a Creamy-bellied Thrush (TWio i wmrochalinus) population from a southern temperate zone in western Argentina. We tound 236 Creamy-bellied Thrush nests of which most were predated (67T). The breeding season was from late October to late December and clutch size was three eggs. Egg survival, hatching success, and fledgling survival of non -depredated nests were quite high <0.6/ - U.U3. 074 ± 0.03. and 0.87 ± 0.04. respectively). The number of eggs in the nest did not affect egg survival or hatching success, bu! number of nestlings in the nest affected fledgling success. Daily nest mortality was higher during the early and late nestling period than durina laving, and early and late incubation periods. Highest nest mortality coincided with periods when activity of parents at' the nest was highest The clutch size was similar to data reported lor thrushes Irom the tropics and south temperate areas, and lower than reported for thrushes from north temperate areas. This latitu ina Pattem ,s similar to the general pattem described for passerines in the tropics and southern temperate areas. Recei\e( < e ruary 1011. Accepted 14 September 201 1. The study of avian breeding ecology in the Americas has mostly concerned tropical and northern temperate species. New data from the Southern Hemisphere suggest life history traits and behavior of southern temperate species are more similar to tropical than to northern temperate species (Martin 1996, Martin et al. 2000, Robinson et al. 2010). This conclusion is based on a limned number of southern temperate locations and species, and more information is needed on avian life history traits in southern temperate areas (Robinson et al. 2010). The Turdidae (true thrushes, Turdus spp.) is a cosmopolitan group in the tropics, and northern and southern temperate areas around Ihe world. Tints, they are an excellent model for study of life history evolution, breeding biology, and latitudi¬ nal variation among related species occupying different ecosystems. However, there are only a comparative studies (Martin et al. 2000, berretti et al. 2005) and the available information ls biased to north temperate species. Most information on breeding biology for species ol T'irdiis in Argentina comes from studies in the Eintpas and Yungas (e.g„ Martin et al. 2000, Suckmann and Reboreda 2003, Ferretti ct al. 2005). There are no studies available tor popula- lions Irom semi-arid areas of western Argentina (but see Mezquida and Marone 2001 for other Passerine species). The Creamy-bellied Thrush ( Turdus amauro- ' Institute Argcntino de Investigaciones de Zonas Aridas. •ADIZA-CONICF.T, Mendoza. Argentina. Corresponding author; e-mai|; aastie @ mendoza-conicet . gob .ar chalinus) is widespread in South America, but its breeding biology has only been recently studied (Astie and Reboreda 2005, 2006). The objectives of our study were to obtain information on breeding biology and reproductive success of a Creamy-bellied Thrush population from a south¬ ern temperate zone in western Argentina. Specif¬ ically, we examined if number of eggs and nestlings present in the nest affect reproductive success during different stages of the nestling cycle. METHODS Study Species.— The Creamy-bellied Thrush is a monomorphic and monotypic passerine ot the genus Turdus. This species is widely distributed from southern Brazil to southern Argentina (Ridge- |y and Tudor 1989), but little is known about its breeding biology. Creamy-bellied Thrushes build open cup nests composed of grasses cemented with mud and lined with grasses and hairs. They lay three esgs with a pale greenish background and spots and blotches of reddish-brown concentrated at the larger pole (Astie and Reboreda 2005). The incubation period is 1 1 .5 days and the nestling period is 12 days (Astie and Reboreda 2005, 2009a). Average adult mass is 55 g (Astie and Reboreda 2005). This species is heavily parasitized by the Shiny Cowbird (Molothms bonariensis) in the study area (Astie and Reboreda 2005, 2006, 2009a, b). Study Site. — The study was conducted at Guay- mallen. Mendoza Province. Argentina (32 51' S, 68 42' W) during the 1999-2002 breeding seasons (Oct-Dec). Mendoza is in the Monte Desert region of Argentina. The Creamy-bellied Thrush only 133 134 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124, No. 1. March 2012 occupies areas irrigated for agriculture. The study area was a 1. OOO-ha cultivated field with vineyards, olives (Olea europaea), and poplar (Papains nigra ) groves. Data Collection. — We followed the fates of 236 thrush nests which were found by observing adult behavior and systematic search. Nests were visited every 1-2 days until nestlings Hedged or the nest (ailed. We recorded the numbers of eggs and nestlings during visits. We considered a nest to have been predated when the complete clutch disappeared between two subsequent visits. Data Analyses. — We divided the reproductive season into groups of 10 days starting with the first egg laid until the last nest in laying stage was found. We recorded the frequency of nests that started in each period and estimated the survival ol nests (0-1) throughout the season with logistic regression including a subset of nests found during building and laying periods (n = 08 nests). We calculated (mean ± SE) success of eggs and nestlings in nests in each nesting cycle period (egg laying, incubation, and nestling) that were not predated, deserted, or parasitized by the Shiny Cowbird. Egg survival was calculated as the proportion of eggs present in the nest at the end of incubation divided by the number of eggs in the nest at the start of incubation. Wc only considered nests found in building and laying stages that survived until the hatch of the first egg („ = 48 nests). We calculated hatching success as the proportion of eggs that hatched divided by the number of eggs in the nest at the end of incubation. We only considered nests found in building, laying, or incubation stages where at least one egg hatched (n = 109 nests). Fledglim- survival was calculated as the proportion of fledglings divided by the number of eggs that hatched. We considered only nests found during building, laying, or incubation stages that fledged at least one nestling (n = 44 nests). We used a Chi-square test to analyze if number of eggs in the nest affected egg survival (all eggs survived or at least one disappeared) or hatching success (all eggs hatched or at least 1 failed), and if fledgling survival was associated with number of nestlings (all nestlings survived or at least I died). We evaluated if nest survival was associated with adult activity in the nest by subdividing the the f8 Cyt ? ,mo tivc Peri°ds: laying (laying of stacel Tavin^ .T"? ^ -VincSon . Ce me laying of the third egg until d .v 7> , incubation ,day S unti, the h« early nestlings (from the day the first egg hatches until day 6). and late nestlings (from day 7 unit the first fledgling left the nest). Nest mortality risk (m) was calculated for each period following Mayfield (1975). and the standard error was calculated as suggested b; Johnson (1979). Wc compared mortality rates with a Fisher test of contingence. We compared each nest mortality period versus every other period and applied a Bonferroni correction (or multiplc comparisons (Abdi 2007). Adult activity in the nest was recorded h videotaping 20 nests with Hi8 Sony vide, cameras. Nests were recorded during 4 hr- beginning at 0700. Cameras were placed 2 m from the nest and camouflaged with leaves We recorded two nests during the laying stage, >u nests during early incubation, two during late incubation, four with early nestlings, and six w-ith late nestlings. We obtained latencies (time elapsed since placement of the camera and the moment m adult returned to the nest ), frequency of visits per hour (average number of times an adult entered the nest in I hr), and nest attentiveness (average proportion of time an adult stayed in the nest during I hr) for each video. All statistical tests were conducted with StatView 5.0 (SAS 1998). RESULTS We found 236 nests of Creamy-bel lied Thrush¬ es. 42% during the building and laying period. 5 1 % during the incubation period, and lac during the nestling period; 18 were found in 1999. 87 in 2000. 91 in 2001. and 40 in 2002. Sixty-seven percent of the nests were predated. 8C7 were deserted, and 2.5% were destroyed by strong winds or rain. Only 22.5% of the nests produced at least one fledgling. The majority of the nests (62%) were parasitized by the Shiny Cowbird and at least one egg in 68% of the nests was punctured. Most of the nests (70%, n = 236) found were in vineyards and in olive trees, and the rest in popL' and fruit trees. Nest dimensions (mean - SC1 were; 10.5 ± 0.5 cm in external height. 6.1 - 0.3 cm in depth, and 12.3 ± 0.2 cm in externa! width with an internal diameter of 7.9 ± 03 cm (n = 20 nests). The first nest was found on 21 October and the number of nests in the laying stage increased to a maximum between 23 and 25 November: the las' nest in the laying stage was found on -- December (Fig. I ). Nest survival was not associ- Astie and Luchesi • CREAMY- BELLIED THRUSH REPRODUCTIVE SUCCESS 135 J3 (A ® C *- 0 L. 0) n E 3 2 Period FIG I. Number of nests in the laying stage during the reproductive season of the Creamy-bellied Thrush. The reproductive season was divided into intervals of 10 days. aied with time in the season when the nest was initialed: nests that failed or succeeded (0- 1 ) were independent of the time when they were initiated (logistic regression: X2 = 1.03, P = 0.31, n = % nests). One egg was laid per day and incubation started witli laying of the penultimate egg (for nests found during building and laying stages, n = 08). tgg survival was 0.67 ± 0.03 (/» 48 nests), hatching success was 0.74 ± 0.03 (n ® 109 nests), and fledgling survival was 0.87 ± 0.04 (n = 44 nests). The number of eggs present in the nest did n'« affect egg survival or hatching success (X2 = m P = 0.09, n = 48 and X2 = 0. 12 ,P = 0.94, 11 ' 109, respectively), hut the number of ffstlings was negatively associated with fledgling success (X2 = 16.35. P = 0.001. n = 44). At least ■he nestling died when there were three nestlings "'58% of the nests (/? = 12 nests), at least one nestling died in 20% of the nests (n = 15 nests) ' lien there were two nestlings, but only 4% died " = 17 nests) when there was one nestling. Bonfenoni correction for multiple com- P^isons set the significance level at y. < 0.005. Dai|.v nest mortality was higher during laying (0-079 ± o.(K)6. n = 39) than during early ,ncubation (0.013 ± 0.002, n = 50, X2, = 8.77, ^ ~ 0.003), but there was no difference with late incubation (0.017 ± 0.002, n = 53, X:i = 4.62, ■D ~ 0.041), early nestling (0.1 1 — 0.006. n = 50, = 2.13. P = 0.1 1), and late nestling (0.10 ± 1-013, n = 24, X2, = 4.40, P = 0.022) periods. We found no differences between early and late incubation (X2 = 1.09. P = 0.27) nor between early and late nestling (X2! = 1.02, P - 0.28) periods. Early incubation was significantly lower than early and late nestling periods (X2, = 37.66. P < 0.001 and X2, = 45.99, P < 0.001, respectively), and late incubation was also significantly lower than early and late nestling periods (X2, = 37.54, P < 0.001. and X2, = 45.14, P < 0.001, respectively; Fig. 2) We had small samples for parental care behavior and could not perform any statistical comparisons. Wc observed a trend of increased latency to return to the nest during the laying stage. Frequency of visits appeared to be highest during the nestling stage and time spent in the nest was highest during the incubation stage (Fig. 3). DISCUSSION The Creamy-bellied Thrush in the southern temperate Monte Desert has several characteris¬ tics typical of tropical birds: low nest survival, high predation rates, and small clutch size. Nests at our study site had a low survival rate, as only 22.5 % of the nests produced at least one lledcling: predation was the major cause ot nest failure (67%). Nest predation was constant throughout the season and was not related to the time a nest was initiated. Nest mortality varied during the nestling cycle and was highest during the late nestling period (Fig. 2). Parental activity during this stage was highest around the nest 136 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 0,14 Laying Early Inc. Late Inc. Early Nest. Late Nest. HG. 2. Daily nest mortality (mean ± SE) during laying, early incubation, late incubation, early nestling ami lute nestling periods of the Creamy-hcllied 1 brush. The number of nests included was 39 in laying, 50 in early incubation. 53 in late incubation, 50 in early nestling, and 24 in late nestling periods. (Fig. 3) and suggests that visual predators were involved. This could be related to attraction of visual predators when parental activity increased around the nest (Skutch 1949, Martin et al. 2000, Martin et al. 2011). An experimental approach would be needed to rule out other alternative hypothesis (e.g., older nesllings/nests may pro¬ duce stronger odors that may attract the attention of predators). We observed that 33% of the eggs disappeared during the incubation stage. This could be caused by partial predation or by brood parasitism, as Shiny Cowbirds puncture eggs in parasitized and in unparasitized nests (Astid and Reboreda 2006 ). Hatching success, and egg and fledgling survival in successful nests (nests that were not predated or deserted) was high (74, 67, and 87%. respectively). Hatching success and egg survival were not related to number of eggs present in tin nest. There was a proportion of eggs that did no hatch (0.26) and this may be caused by infertilit\ or by death of the embryo caused by insufficien incubation (Davies and Brooke 1988. Svenssoi et al. 2007). We found that number of egg: present in the nest had no effect on hatching success and it is unlikely that insufficien incubation was the main cause. Nestlings dice rom starvation more often when they shared the “rr lhT °ne Sib,ing- siting that food competition between nestlings was strong Most of the nestlings survived when only one or two eggs hatched due to previous egg loss or hatching failure. However, when all eggs hatched, at least one nestling died. The brood reduction hypothesis (Lack 1947, Ricklefs 1965) suggests that some species lay more eggs than they can rear nestlings because food supply may vary unpre- dictably. Thus, parents may lay a large clutch appropriate for a food-rich year and. in food-poor years, nestling survival would be mediated by sibling competition. The brood reduction hypoth¬ esis could be a possible explanation for Creamy- hcllied Thrush nestling mortality. However, future studies should evaluate if nests with three nestlings improve nestling survival in food-good years. The clutch size of Creamy-bellied Thrushes at our study site was small (mean = 3 eggs) and typical of tropical and southern temperate birds There is a lack of information of the breeding biology of most thrushes, but the available data suggests this group has the same pattern as other passerines. The clutch size reported for thrusto' from north temperate areas is 4-5 eggs (Scbnao- 1991, Arheimcr and Svensson 2008. Morton am- Pereyra 2010), and is 2-3 eggs for thrushes a tropic and south temperate areas (Lichtensteu 1998. Sackmann and Reboreda 2003. Akinp- ■' 2005, Astie and Reboreda 2005, Halupka and Greeney 2009). Ours is the first study presenting data from a Tardus species that inhabits an so temperate area as far as we know. We did not Aslie and Luchesi • CREAMY-BELLIED THRUSH REPRODUCTIVE SUCCESS 137 I ~.J 8 7 6 5 4 3 2 1 0 FIG. 3. Latency (time in seconds an adult spent in returning to the nest after we placed the camera), frequency (number of visiis/hr). and nest attentiveness (percentage of time in the nest during each penod). All data are presented mean _ . conduct our study in a tropical environment, but life-history traits of the Creamy-bellied Thrush are similar to tropical birds. Future research on the breeding biology of southern temperate thrushes may offer new understanding of the evolution of life-history traits. ACKNOWLEDGMENTS We thank P. E. Llambias and F. Fernandez Campon for helpful comments on this manuscript. NL was supported by a fellowship from Consejo Nacional de Investigaciones Cientificas y Tecnicas (CONICET). AA is a Research Fellow of CONICET. 138 THE WILSON JOURNAL OF ORNITHOLOGY • Vol 124. No. 1. March 2012 LITEREATURE CITED Abdi. II. 2007. Bonferroni and Sidak corrections for multiple comparisons. Encyclopedia of measurement and statistics Salkind Edition. Thousand Oaks. California, USA. AKINPLU , A. I. 2005. A twelve-month field study of the West African Thrush Tardus pelios (Passeriformes: Muscicapidae). Part 2. Annua] cycles. Revista de Biologia Tropical 53: 239-247 Arheimlk. O. and S. Svensson. 2008. Breeding perfor¬ mance of the Fieldfare Tardus pilaris in the subalpine birch zone in southern Lapland: a 20 year study. Ornis Svecica 18:17^14. Asti£, A. A. and J. C. Rebokeoa. 2005. Creamy-bellied Thrush defenses against Shiny Cowbird brood para¬ sitism. Condor 107:788-7%. AstiE. a. A. AND J. C. Reborepa. 2006. The cost of egg punctures and parasitism by Shiny Cowbirds (Moln- tlirus honarienxis) at Creamy-bellied Thrush ( Tardus auiuurochtdinus) nests. Auk 123:23-32. AsuE, A. A. and J. C . Reborepa. 2009a. Function erf egg punctures by Shiny Cowbirds in parasitized and nonpnrasiti/ed Creamy-bellied Thrush nests. Journal of Field Ornithology 80:336-343. AstiE, A. A. and .). C. Reborepa. 2009b. Shiny Cowbird parasitism ol a low quality host: effect of host trails on a parasite's reproductive success. Journal of Field Ornithology 80:224 233. Davies. N. B. and M. de L. Brooke. 1988, Cuckoos versus Red Warblers: adaptations and counteradaptu- tions. Animal Behaviour 36:262-284. Ferretti, V.. p. E, Llambias. and T. E. Martin. 2005. Life-history variation of a neotropical thrush challeng¬ es rood limitation theory. Proceedings of the Royal Society of London. Series B 272:769-77.3. Halupka. K. and H. F. Greeney. 2009. Breeding biology ol Pale-eyed Thrushes (Tardus kttcops) in the cloud forest of northeastern Ecuador. Omitologia Neotrop¬ ical 20:381-389. Johnson. D. H. 1979. Estimating nest success: the Mayfield method and an alternative. Auk 96:651-661. Lack. D. 1947. The significance of clutch size. Ibis 89:302-352. Lichtenstein. G. 1998. Parasitism by Shiny Cowbird of Rufous-bellied Thrushes. Condor 100:680-687. Martin, T. E. 1996. Life history evolution in tropical and south temperate birds: what do we really know? Journal of Avian Biology 27:263-272. Martin. T. E.. P. R. Martin, C. R. Oison. b. J. Hfjiknwk and .1. .1. Fontaine. 20tX), Parental care and clutch sizes in North and South American birds. Science 2X7 : 1 4&2- ! 4«5 Martin, T. E., P. Li.oyo, C. Bosque. D. C. Barton, a I. Biancucci. Y. Cheng, and R. Ton. 201 1 Growil> rate variation among passerine species in tropical and temperate sites: an antagonistic interaction between parental food provisioning and nest predation mk. Evolution 65:1607-1622. Mayfield, H. F. 1975. Suggestions for calculating not success. Wilson Bulletin 87:456-466. Mezquida. I-', T. AND L. Marone. 2001. Factors affecting nesting success of a bird assembly in the central Monte Desert. Argentina. Journal of Avian Biology 32 287-296 Morton, M. L. and M. E. Pereyra. 2010. Dnelopnaeni of incubation temperature and behavior in thrushes nesting at high altitude. Wilson Journal of Ornithology 12266644 RK. Kt.bFS. R. E. 1965. Brood reduction in the Curse- Mid Thrasher. Condor 67:505-510. Ridgely, R. S. and G. Tudor. 1989, The birds of South America, flic oscine passerines. Oxford University Press, New York. USA. Robinson. D„ M. Hau, K. C. Klasing, M. Wikelski. J D. Brawn. S. II. Austin, C. E. Tarwater, and R E. Ric ki kes. 2010. Diversification of life histories in New World birds. Auk 127:253-262. Sac kmann. P. and J. C. Reboreda. 2003. A comparative study of Shiny Cowbird parasitism of two large host', the Chalk -browed Mockingbird and the Rufus- bellied Thrush. Condor 105:728-736. •SAS Institute Inc. 1998. StatView user's guide 5 0. SAS Institute Inc.. Cary, North Carolina. USA. Si knack. S. 1991. The breeding biology and nestling diet ol the Blackbird Turdux merula L. and the Song Thrush Tardus philometos C. L. Brehni in Vienna an: in an adjacent wood. Acta Omithologica 26:B5- 1 05 Ski i ten. A. F 1949. Do tropical hirds rear as it.it hit 1 they can nourish? Ibis 91:430-455. Svensson. M.. P. T. Rintamaki. T. R. Birkhead. S. C Griffith, and A. Lundberg. 2007. Impaired hatching success and male-biased embryo mortality in Tree Sparrows. Journal of Ornithology 148:1 17-122. The Wilson Journal of Ornithology 124(1): 139 145. 2012 FLANGE COLOR DIFFERENCES OF BROOD PARASITIC BROWN-HEADED COWBIRDS FROM NESTS OF TWO HOST SPECIES REBECCA CROSTON.' 7 K CHRISTOPHER M. TONRA,2-5 SACHA K. HEATH,2,36 AND MARK E. HAUBER1 4 ABSTRACT— We compared the red. green, and blue color values from digital photographs of the rictal flanges of nestling Brown-headed Cow-birds (Molothrus ater). a generalist obligate brood parasite, in sympatric Yellow Warbler iSetopliaga petechia) and Song Sparrow ( Meloxpizu mclojial nests at Mono Lake. C ulifornia. USA. We detected significant differences in all three color components across nestlings of different species (R: P < U.0001 : G: P < 0.00131. B . P < 0.0001 ). hut differences among cowbird nestlings from the nests of these two hosts were not significant (R : P — 0.543. G. P = 0.737: B: P = 0.319). Principal components results were mixed: Principal Component I described brightness and accounted for 84% of the variance. It did not differ among cowbird nestlings front nests of different hosts (P — 0.319). Principal Component IT described chromaiicity and accounted for \4% of the variance, which differed significantly among cowbird nestlings from the two different hosts' nests I P = 0.026). Color differences between cowbird nestlings from nests of different host species may result from selective parasitism by female parasites based on host nestling flange morphology, or ontogenetic effects on cowbird nestlings reared by different host species. Received 25 January 2011. Accepted 21 July 2011. Evidence of recognition and discrimination of parasitic nestlings is relatively rare among hosts of avian brood parasite species (Redondo 1993; Grim ct al. 2003; Langmore et al. 2003. 2009; Schuetz 2005b; Sato et al. 2010; Shizuka and Lyon 2010). Patterns of parasite chick's visual and/or acoustic similarity of host nestlings in a handful of brood parasite lineages ( Anderson et al. 2009, Sato et al, 2010, Langmore cl al. 2011) imply mimicry to avoid rejection (Langmore ct al. 2003, Payne 2005, Tokue and Ueda 2010, Langmore et al. 2011). Hosts may discriminate not only by directly rejecting foreign nestlings, but by providing belter care or higher quality prey (Schuetz 2005a. Solcr 2008) for nestlings with particular attributes (Rothstein 1978, Lichtenstein 2001. Dugas 2009), resulting in variation among nestlings in growth rale and condition (Hauher Graduate Program in Biology. Ecology. Evolutionary biology and Behavior Subprogram, Graduate Center. City University of New York, NY 10016. USA Department of Wildlife. Humboldt Slate University, Areata. CA 95521. USA. PR BO Conservation Science. Petaluma, CA 94954, USA. 'Department of Psychology, Hunter College. City University of New York, NY 10065, USA. Current address: Smithsonian Conservation Biology Institute. National Zoological Park. P. O. Box 37012-MRC 5503, Washington, D C. 20013. USA. Current address: Department of Wildlife. Humboldt Stale University. Areata. C'A 95521. L SA. Department of Psychology. Hunter College. 695 Park Avenue. New York. NY 10065. USA. Corresponding author; e-mail: RCroston@gc.cuny.edu and Kilner 2007). Nestling discrimination among cowbird hosts has been documented for Rufous- bellied Thrushes ( Turdus rufiventris ) parasitized by the non-evicting generalist Shiny Cowbird (Molothrus bonuriensis ) (Lichtenstein 2001), but is not yet known to occur in any hosts of the Brown-headed Cowbird (M. titer). Hosts may recognize parasitic nestlings using variation in size, color, vocalization, brood size, and length of time before fledging (Langmore et al. 2003. Schuetz 2005b, Grim 2007). Variable coloration of both gapes and rictal flanges may have a signaling function, conveying nestling identity, need, health or other indicators of quality (Thorogood et al. 2008, Dugas 2010), which would need to be matched by parasitic nestlings (Nicolai 1974, Payne 2005, Hauber and Kilner 2007). Flange color is typically monomorphic within species, but Brown -headed Cowbird nest¬ lings appear polymorphic across the species so that a nestling has either distinctly yellow or white flanges with lew intermediates (Rothstein 1978). Polymorphic flange color occurs in only two phylogenetically distant New World oscine gen¬ era. Geospiza and Molothrus. It is plausible this polymorphism in cowbirds is the outcome ot selection for preferential parasitism of certain host species by female cowbirds to match the host- specific flange color by the parasitic nestling (Ellison et al. 2007). Alternatively, the differences in human-per¬ ceived flange phenotype ol cowbird nestlings may not result from genetic polymorphism, but from differences in carotenoid consumption and 139 140 THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 1. March 2012 provisioning to chicks across different hosts. Carotenoid pigments are derived entirely from diet, and their concentration is known to modulate nestling mouth color (Thorogood et al. 2008). Carotenoid concentration is widely hypothesized to indicate nestling quality, as demonstrated in House Sparrows ( Passer domesticus) (Loiseau et al. 2008) and Bam Swallows ( Hirundo rusticci ) (Saino et al. 2000, 2003). We investigated whether cowbird nestlings differ in flange coloration when reared by one of two host species. Song Sparrows (Melospiza melodia) and Yellow Warblers (Setophaga pete¬ chia) using quantitative measures of coloration. Birds have a fourth violet- or ultraviolet-sensitive photoreceptor type, and human color perception is an insufficient proxy for avian color perception (Cuthill et al. 2000). We provide the first objective assessment of cowbird nestling flange colors based on measures of color using digital photographs and imaging software (Dale 2000). However, this remains a preliminary analysis because imaging software is designed for human vision and has limited value for avian perceptual studies (Stevens et al. 2007). The objective of our study was to test the hypothesis that rictal flange color of host and cowbird nestlings varies between nestlings and parasites of two sympatric hosts. We predicted measures of flange colors would differ among: ( I ) nestlings of different species (Yellow Warbler, Song Sparrow, and Brown-headed Cowbird), and (2) cowbird nestlings in nests of different host species. METHODS Study Site and Species. — This study was con ducted in the riparian corridors of four tributarie of Mono Lake (38° 1’ N, 119° 3' W) on th- eastern slope of the Sierra Nevada, California USA: Lee Vining, Mill, Rush, and Wilson creeks We located Song Sparrow' and Yellow Warble nests and monitored nests during the 200< breeding season lollowing Martin and Geupe (1993) and Ralph et al. ( 1993). The ranges of nes initiation dates for Yellow Warbler and Sonj Sparrows were similar in 2004 (Tonra et al. 2009) Nestling Photographs,- We photographed nest mgs on day 6 (hatching day = day 0) of the cowbird nestling cycle. This day was chose, because .t coincided with the age at which allowed f ^ S"tTicien,i>' ***& to band and allowed for incorporation of carotenoid pigments from host-provisioned diet into tissues. Photo¬ graphs were taken with a Hewlett Packard Photosmart 215 digital camera, set to ISO 200 and 'fine' quality (1,280 X 960 pixels). We photographed all cowbirds in each nest and. if hosts were present, we also randomly selected one individual to photograph. Sixteen Brown-headed Cowbird. three Yellow Warbler, and three Song Sparrow nestlings were included in the analysis. Both cowbird and host chicks were photographed in two Song Sparrow and two Yellow Warbler nests, and one Yellow Warbler and two Song Sparrow nests each contained two cowbird nestlings. Each photograph was taken of the nght side of the head against a background of gray paper with a strip composed of six 1-cnr sections cut from paint store color sample cards (red. blue, green, white, yellow, and black) (Home Depot, Reno. NV, USA) as a color standard, and stored in a dark box between photography sessions. This allowed us to make direct comparisons of colors under varying light conditions in the field. The photographs were saved and subsequently ana lyzcd as jpeg images using the histogram function in Adobe Photoshop Elements 8.0 (Adobe Sys¬ tems Inc.. San Jose. CA, USA). Storing images as jpeg compresses both image and color data (Stevens et al. 2007); color compression in this software obscures rather than enhances differenc¬ es in color and would result in our failure to reject the null hypothesis (despite its falsehood: Type II statistical error). Thus, use of jpeg images made our analyses more conservative. Flanges were divided into three portions tor color measurement: A at the apex of the flange, site B at the fleshy middle, and site C at its most rostral point. Three replicate measures of red. green, and blue values were made at each flange site for each nestling. Red, green, and blue measures represent the intensity levels (satura¬ tion) for 24-bit color; these measures range in intensity from 0 (black) to 255 (white, totally saturated color). Measurements were also made from the center of the yellow standard present in each photograph to allow for direct comparison of the standard and biological colors. Data Analysis. — We compared red (R). green (G), and blue (B) values, while accounting lor variation in light conditions associated with held work in each photograph, by first scaling according to the RGB values of the yellow standard in that photograph i. Thus, F(A| * where I = R. G, or B value, F, = flange color Croston el al. • FLANGE COLOR DIFFERENCES OF COWBIRDS 141 ll II 1 1 ll SOSP YWAR BHCO in SOSP BHCO in YWAR nest nest FIG. 1. Differences in flange color saturation values by species and host. SOSP Song Sparrow, YWAR Yellow Warbler, and BHCO = Brown-headed Cowbird. Bars represent mean saturation values of red. green, and blue across flange sites and replicates. Error bars represent standard error, value for i, and Y| = the yellow standard for i. Repeated measures at each flange site within color group (R, G, and B) were averaged across replicates. Data were analyzed using three sepa¬ rate univariate analyses of variance (Mixed Effects ANOVA), where each color value was the dependent variable, nestling species (Yellow Warbler, Song Sparrow, or Brown- headed Cow- bird) and cowbird nestling host species (Yellow Warbler or Song Sparrow) were fixed effects, and nestling metal-band ID and flange site were random effects. Color variables R, G, and B are necessarily correlated and we also used principal components analysis (PCA) to recombine color variables into uncorrelated scores describing brightness and chromaticity (following Endler andThery 1996). Mean PC scores were compared among host nestling species and cowbird nestling species groups using ANOVA. All analyses were conducted in JMP Version 8.0 (SAS Institute Inc.. Cary, NC, USA). RESULTS Color values were significantly different among nestling species in all three univariate analyses of red. green, and blue color components (R: Fz.ax = 1 * .67, P < 0.000 1 : G: F2.w = 1 1 - 1 4, /* < 0.000 1 ; B: *2.50 = 16.61, P < 0.0001), but not between cowbird nestlings of different host species