Raptor Research Volume^ 37 Number 4 December 2003 ^ JETpSfVi'-^^ ^ * i_Elll'l ■TW IMjrIkS^u^dF^ iJ: LiMjai \ tfj r £ 1 K '■ rJQK** 1 ■ ■* Published by The Raptor Research Foundation, Inc. THE RAPTOR RESEARCH EOUNDATION, INC. (Founded 1966) OFFICERS PRESIDENT: Brian A. Millsap SECRETARY: Judith Henckel VICE-PRESIDENT: David M. Bird TREASURER: Jim Fitzpatrick BOARD NORTH AMERICAN DIRECTOR #1; Jeff Smith NORTH AMERICAN DIRECTOR #2: Laurie J. Goodrich NORTH AMERICAN DIRECTOR #3: Ted Swem INTERNATIONAL DIRECTOR #1: Beatriz Arroyo INTERNATIONAL DIRECTOR #2: Ricardo Rodriguez-Estrella Hi DIRECTORS INTERNATIONAL DIRECTOR #3: Steve Redpath DIRECTOR AT LARGE #1: Jemima ParryJones DIRECTOR AT LARGE #2: Petra Bohalu Wood DIRECTOR AT LARGE #3: Michael W. Collopy DIRECTOR AT LARGE #4: Carol McIntyre DIRECTOR AT LARGE #5: Robert N, Roseneield DIRECTOR AT LARGE #6; Daniel E. Varland EDITORIAL STAFF EDITOR: James C. Bednarz, Department of Biological Sciences, P.O. Box 599, Arkansas State University, State University, AR 72467 U.S.A. ASSISTANT EDITOR: Rebecca S. Maul ASSOCIATE EDITORS James R. Belthoff Juan Jose Negro Clint W. Boal Marco Restani Michael I. Goldstein Fabrizio Sergio Joan L. Morrison Ian G. Warkentin BOOK REVIEW EDITOR: Jeffrey S. Marks, Montana Cooperative Research Unit, University of Montana, Missoula, MT 59812 U.S.A. SPANISH EDITOR: Cesar Marquez Reyes, Institute Humboldt, Colombia, AA. 094766, Bogota 8, Colombia EDITORIAL ASSISTANTS: Jaimie Varner, Joan Clark The Journal of Raptor Research is distributed quarterly to all current members. Original manuscripts dealing with the biology and conservation of diurnal and nocturnal birds of prey are welcomed from throughout the world, but must be written in English. Submissions can be in the form of research articles, short communications, letters to the editor, and book reviews. Contributors should submit a typewritten original and three copies to the Editor. All submissions must be typewritten and double-spaced on one side of 216 X 278 mm {Wz X 11 in.) or standard international, white, bond paper, with 25 mm (1 in.) mar- gins. The cover page should contain a title, the author’s full name(s) and address (es). Name and address should be centered on the cover page. If the current address is different, indicate this via a footnote. A short version of the title, not exceeding 35 characters, should be provided for a running head. An abstract of about 250 words should accompany all research articles on a separate page. Tables, one to a page, should be double-spaced throughout and be assigned consecutive Arabic numer- als. Collect all figure legends on a separate page. Each illustration should be centered on a single page and be no smaller than final size and no larger than twice final size. The name of the author(s) and figure number, assigned consecutively using Arabic numerals, should be pencilled on the back of each figure. Names for birds should follow the A.O.U. Checklist of North American Birds (7th ed., 1998) or another authoritative source for other regions. Subspecific identification should be cited only when pertinent to the material presented. Metric units should be used for all measurements. Use the 24-hour clock (e.g., 0830 H and 2030 H) and “continental” dating (e.g., 1 January 1999). Refer to a recent issue of the journal for details in format. Explicit instructions and publication policy are outlined in “Information for contributors,”/. Raptor Res., Vol. 37(4), and are available from the editor. Submit manuscripts to J. Bednarz at the address listed above. COVER: Northern Barred Owl {Strix varia varia) and northern Spotted Owl {S. occidentalis caurina). Painting by Patty Walcott, P.O. Box 545, Tenino, WA 98589 U.S.A. (e-mail: sunnmoon@scattercreek.com) Contents Distribution, Numbers, and Site Characteristics of Spotted Owls and Barred Owls in the Cascade Mountains of Washington. Robert R. Pearson and Kent B. Livezey 265 Temporal and Spatial Stabiliiy of Red-tailed Hawk Territories in the Luquillo Experimental Forest, Puerto Rico, cunt w. Boai, Helen a. Snyder, Brent d. Bibles, and Tracy S. Estabrook 277 Wintering Raptor Use of Hybrid Poplar Plantations in Northeastern Oregon. Brian W. Moser and G. Keith Hilpp 286 Mortality of Wintering Ospreys and Other Birds at Aquaculture Facilities in Colombia. Marc J. Bechard and Cesar Marquez-Reyes 292 Blood Chemistry, Cytology, and Body Condition in Adult Northern Goshawks {ACCIPITER GENTILIS). Lauri A. Hanauska-Brown, Alfred M. Dufty, Jr., and GaryJ. Roloff. 299 Productivity and Population Trends of Ospreys in the Kawartha Lakes Region, Ontario, 1978—2001. Shane R. de Solla, Pamela A. Martin, Peter J. Ewins, and Michael E. Barker 307 Habitat Characteristics of Great Gray Owl Sites in the Siskiyou Mountains of Southwestern Oregon. Trevor W. Fetz, Stewart W. Janes, and Heidi Lauchstedt 315 Nest-site Characteristics and Linear Abundance of Cliff-nesting American Kestrels on San Clemente Island, California. Brian L. Sullivan, Eric l. Kershner, Sean P. Finn, Anne M. Condon, Douglass M. Cooper, and David K Garcelon 32.3 Short Communications Cuban Raptor-migration Counts in 2001. Freddy Rodriguez Santana, Luis Meii^ Hernandez, Mark Martell, and Keith L. Bildstein 330 Territory Change and Nest-site Switching in the Bearded Vulture (Gypaetus BARBATUS). Antoni Margalida, Jordi Canut, and Diego Garcia 333 Immature Northern Goshawk Captures, Kills, and Feeds on Adult-sized Wild Turkey. Gregory H. Golet, Henry T. Golet, and Albie M. Colton 337 A Helper at the Nest of Peregrine Falcons in Northern Japan. Takashi Kurosawa and Reiko Kurosawa 340 Two Cases of Cooperative Breeding in Eurasian Hobbies. Inigo Zuberogoitia, Jose Antonio Martinez, Ainara Azkona, Agurtzane Iraeta, Ihaki Castillo, Raul Alonso, and Sonia Hidalgo 342 Food Habits of Peregrine Falcons in Kentucky. Kristina m. Carter, Michael j. Lacki, Matthew R. Dzialak, Laura S. Burford, and Ronald O. Bethany 344 Letter Harris’s Hawk Dies After Colliding with Motor Vehicle. Jennifer O. Coulson and Thomas D. Coulson 350 Book Review. Edited by Jeffrey S. Marks 352 Information for Contributors 354 Index to Volume 37 358 The Raptor Research Foundation, Inc. gratefully acknowledges funds and logistical support provided by Arkansas State University to assist in the publication of the journal. THE JOURNAL OF RAPTOR RESEARCH A QUARTERLY PUBLICATION OF THE RAPTOR RESEARCH FOUNDATION, INC. VoL. 37 December 2003 No. 4 J Raptor Res. 37(4):265-276 © 2003 The Raptor Research Foundation, Inc. DISTRIBUTION, NUMBERS, AND SITE CHARACTERISTICS OE SPOTTED OWLS AND BARRED OWLS IN THE CASCADE MOUNTAINS OF WASHINGTON^ Robert R. Pearson 159 Alpine Drive, Packwood, WA 98361 U.S.A. Kent B. Livezey^ U.S. Fish and Wildlife Service, Western Washington Fish and Wildlife Office, 510 Desmond Drive, Lacey, WA 98503 U.S.A. Abstract. — ^We analyzed distribution, numbers, and site characteristics of northern Spotted Owls {Strix occidentalis caurina) and northern Barred Owls {Strix varia varia) in Gifford Pinchot National Forest, Washington from July 1978-November 2001. Spotted Owl site-centers averaged significantly higher in elevation and in areas with steeper slopes than Barred Owl site-centers. Relative percentage of Barred Owl detections increased 8.6% annually during the study period. Plots of 0.8-km radius (201 ha) cen- tered on Spotted and Barred owl sites differed significantly from random plots {N = 300); plots of both species contained more forest ^180 yr old and fewer ha of forest 50-79 yr old, and Spotted Owl plots contained more forest 130-179 yr old. Spotted Owl site-centers were in significantly older forest than Barred Owl and random site-centers. Occupied Spotted Owl sites with timber harvest since 1978 con- tained significantly more forest ^80 yr old than unoccupied sites with harvest. Occupancy of Spotted Owl sites in 2001 relative to number of Barred Owl sites, and slope and elevation of Spotted Owl sites, could be traced directly or indirectly to the presence of Barred Owls. Key Words: northern Barred Owl, Strix varia varia; northern Spotted Owl, Strix occidentalis caurina; occu- pancy, site characteristics-, Washington-, Cascade Mountains. DISTRIBUCION, NUMEROS Y GARACTERISTICAS DE SITIOS DE STRIX OCCIDENTALIS Y STRIX VARIA EN LAS MONTANAS CASCADES DE WASHINGTON Resumen. — Analizamos la distribucion, los numeros, y las caracteristicas de los sitios de Strix occidentalis caurina, considerados en peligro de extincion en los Estados Unidos, y de Strix varia varia, que estan invadiendo el noroeste de los Estados Unidos. El area de estudio abarco 217 812 ha en el Bosque Nacional Gifford Pinchot, Washington, y estudiamos estos buhos desde julio de 1978 a noviembre del 2001. En promedio, los sitios de Strix occidentalis estaban en lugares significativamente mas altos en elevacion y mas inclinados que los de Strix varia. El porcentaje relativo de detecciones de Strix varia aumento 8.6% al aho. Los sitios de un radio de 0.8 km (201 ha) de Strix occidentalis y las de Strix varia difirieron de los sitios al azar {N = 500) porque los sitios de ambas especies tenian mas areas de bosques de por lo menos 180 anos de edad y menos areas de bosques de 50-79 ahos de edad, y los sitios de Strix occidentalis tenian mas areas de 130—179 ahos de edad. Los centros de los sitios de Strix occidentalis estaban situados en bosques mas viejos que los de Strix varia y los de sitios al azar. Los sitios ocupados por Strix occidentalis con tala de arboles desde 1978 tenian mas areas de bosques de por lo menos 80 ahos de edad que los sitios inocupados con tala de arboles. La ocupacion de los sitios de Strix occidentalis ^ The views herein reflect those of the authors and are not necessarily those of the U.S. Fish and Wildlife Service ^ Corresponding author’s e-mail address: kentJivezey@fws.gov 265 266 Pearson and Livezey VoL. 37, No. 4 en 2001 en relacion al numero de los sitios de Strix varia, y la inclinacion y elevacion de los sitios de Strix occidentalis, puede estar directamente o indirectamente relacionada a la presencia de Strix varia. [Traduccion de Kent Livezey] The northern Spotted Owl (Strix occidentalis cau- nna) is a resident of forest areas in southwestern British Columbia, western Washington and Oregon, and northwestern California. It was listed as a “threatened” subspecies in 1990, primarily due to evidence that the population was declining concurrent with loss of habitat (Federal Register 55:26114—26194, 26 June 1990, Gutierrez 1994). The Barred Owl (Strix varia) was originally wide- spread throughout southeastern Canada, eastern United States, and eastern Mexico (Rignall 1973). In the early 1900s, the northern Barred Owl (S. v. varia) began to expand is range westward, moving across southern Canada and south into Washing- ton, Oregon, and California. Barred Owls were first detected in Washington in 1965 (Rogers 1966), Oregon in 1974 (Taylor and Forsman 1976), and California in 1981 (Evens and Le Valley 1982). The range of the Barred Owl now nearly completely overlaps that of the northern Spotted Owl. Barred Owls are larger than Spotted Owls, are aggressive toward them (e.g., Hamer et al. 1989, Leskiw and Gutierrez 1998), and may compete with them (Hamer et al. 1989, Herter and Hicks 2000, Kelly et al. 2003). Although a number of authors have suggested that Barred Owls compete with Spotted Owls, the range expansion of the Barred Owl is so recent that there have been only a few studies in which rela- tionships between Barred Owls and Spotted Owls have been investigated (Hamer 1988, Hamer et al. 1989, 1994, 2001, Herter and Hicks 2000, Kelly 2001, Kelly et al. 2003). In this paper, we present data from a 24-yr study (1978-2001) during which we monitored the distribution, number, and habitat associations of Barred and Spotted owls in south- western Washington. Our objectives were to: (1) de- scribe temporal changes in the relative number and distribution of Barred and Spotted owls; (2) com- pare habitats occupied by both species, and (3) eval- uate occupancy of territories by Spotted Owls as a function of the presence or absence of Barred Owls. Methods Study Area. The study area was the Cowlitz Valley Rang- er District of the Gifford Pinchot National Forest (GPNF). This 217 812-ha area was located on the west slope of the Cascade Mountains in southwestern Wash- Figure 1. Location of study area (shaded area) in south- ern Washington. ington (Fig. 1). Forested vegetation on the study area was dominated by Douglas-fir (Pseudotsuga menziesii) , western hemlock (Tsuga heterophylla) , mountain hemlock (Tsuga mertensiana) , Pacific silver fir (Abies amabilis), noble fir (Abies procera), subalpine fir (Abies lasiocarpa), and west- ern redcedar ( Thuja plicata) . Small, local areas of decid- uous hardwoods (1146 ha), such as bigleaf maple (Acer macrophyllum) , red alder (Alnus rubra) , and black cotton- wood (Populus trichocarpa) , were mostly in riparian areas Much of the study area exhibited the effects of large- scale fires, resulting in large blocks of contiguous forest of the same general age. Forest harvest began in the study area in about 1947. By 2001, forested areas 50-129 yr old were still relatively untouched by timber harvest, whereas forested areas ^130 yr old either were heavily fragmented by logging outside of wilderness or remained in contiguous blocks within wilderness. In some areas, older forest was found in narrow bands along valley bot- toms, intergrading into younger forest on upper slopes and ridges. The four main forest zones in the study area, with approximate elevation limits, were western hemlock (<914 m), Pacific silver fir (914-1373 m), mountain hemlock (1372—1708 m), and subalpine fir (1707 m-tree- line) (GPNF unpubl. data). The Douglas-fir, a pioneer species that repopulates areas denuded by forest fire, was present throughout the study area within all four forest zones. In addition, old-growth Douglas-fir were present throughout the study area as remnants left after forest fires. In 2001, the study area was composed of ca. 52% forest S80 yr old, 33% forest <80 yr old, and 15% non- December 2003 Spotted Owls and Barred Owi.s 267 forested areas (rock, wet-mesic, dry meadow/brush, wa- ter). The study area contained ca. 2630 km of roads. The Northwest Forest Plan (USDA and USDI 1994a, 1994b) placed every ha of federally-administered land within the range of the northern Spotted Owl into one of many land-use allocations (LUAs). Within the study area, the LUAs in which timber harvest was permitted were matrix (49 716 ha) and an adaptive management area (42 741 ha), and the LUAs in which no timber har- vest was permitted were late-successional reserves (55 275 ha), congressionally-reserved areas (42 721 ha), and ad- ministratively-withdrawn areas (27 359 ha). The North- west Forest Plan permits certain forest activities in late- successional reserves, such as thinning in stands ^80 yr old, when they are beneficial to the creation of late-suc- cessional forest characteristics. Critical habitat units (Fed- eral Register 57:1796-1838 15 Jan 1992) (81597 ha) overlaid all other LUAs except congressionally-reserved areas; timber harvest was permitted in critical habitat if it overlaid an adaptive management area or matrix. Survey Methods. Spotted and Barred owls respond to imitations of their calls, and Barred Owls readily respond to Spotted Owl calls (Forsman et al. 1984, McGarigal and Fraser 1985, Flamer 1988, Reid et al. 1999). Amplified tape recordings or voice imitations of Spotted Owl calls were used to detect Spotted Owls and Barred Owls. Hy- brid Spotted X Barred owls were identified by their dis- tinctive five- or six-note calls that are “strikingly differ- ent” from Spotted Owl four-note calls and Barred Owl eight-note calls (Hamer et al. 1994). Surveys followed ac- cepted protocols (Forsman 1983). Spotted Owl habitat was identified within each survey area and survey stations were located to achieve complete coverage of the habitat. Survey stations were 0.2-0. 8 km apart within Spotted Owl habitat, typically along roads. Each survey was conducted for at least 10 min per station unless a Spotted Owl, Barred Owl, or Great Horned Owl {Bubo virginianus) was detected, in which case calling was stopped to minimize disturbance to Spotted Owls and potential confrontations between owl species. Survey effort varied from year to year. Forest Service surveys were conducted according to several region-wide survey programs, including random-area surveys and sur- veys surrounding proposed timber sales, targeting por- tions of the entire study area each year with no single year achieving total coverage. However, by 1992, the en- tire study area had been surveyed cumulatively. From 1994-2001, R. Pearson achieved nearly complete cover- age in even years and in 2001, while the other years had partial coverage. Because of this, our total sites for Spot- ted and Barred owls represent cumulative totals rather than the total in 2001. Data Analysis. In our analysis, a Spotted or Barred owl “site” was analogous to an individual owl territory. Sites were designated based on the presence of nests, fledged young, or repeated detections of paired or single owls in the same general area. Resident single sites required s three detections of a male or female owl on three differ- ent survey outings. Pair sites required ^ one detection of both a male and a female that was >1.6 km from a known site for Spotted Owls and >0.8 km from a known site for Barred Owls. The 1.6-km requirement for Spotted Owl sites was waived in one instance when a new pair site with fledglings was identified <1.6 km from two known sites, and adults of the adjacent pairs were also detected. For Barred Owls, we also used a category called “poten- tial sites” that included suspected territories that were based on one or two detections of a single owl in an area that was ^0.8 km from any other Barred Owl sites. The 0.8-km cutoff for designation of potential sites was based on Hamer et al. (1989), who reported a 218-ha mean sum- mer home-range for Barred Owls (a circular plot of 0.8- km radius includes 201 ha). For each site, we plotted a “site-center” based on the location of the nest tree, fledged young, multiple detections of a pair, or multiple detections of single owls. For Spotted Owl pairs that changed nest trees or centers of activity within sites, we used the most recent locations to plot site-centers. For comparison of forest-stand age at the site-center, we iden- tified the age of the forest stand that included the site- center. We considered Spotted Owl sites unoccupied if they were occupied by a pair for 2:1 yr, and were subse- quently surveyed 2:10 times during the last 5 yr of the study (1997-2001) with no detections of Spotted Owls within a 1.6-km radius. We excluded sites that were not surveyed >10 times during the last 5 yr of the study {N = 16) from analyses of Spotted Owl site occupancy. We an- alyzed landscape attributes around site-centers based on comparisons of vegetation composition within a 0.8-km ra- dius centered on the owl site-centers and random loca- tions, which we called “core-plots” (Hunter et al. 1995, Meyer et al. 1998, Swindle et al. 1999, Herter and Hicks 2000, Kelly 2001). We centered random sites on the first 500 randomly generated, terrestrial locations below 1524 m in elevation. We excluded random locations above 1524 m elevation because we detected no Spotted or Barred owls above 1524 m, and excluded non-forested areas (rock, wet-mesic, dry meadow/brush, and water) from veg- etation analyses. To approximate Spotted Owl home rang- es in Fig. 2, we used a 2.9-km radius circle (2670 ha) fol- lowing Herter and Hicks (2000). For each owl site, we determined a “nearest-neighbor distance” which was the distance to the nearest known site-center of the same spe- cies (Hamer et al. 1989, Anthony 2001). We estimated the “minimum nearest-neighbor distance” for Spotted Owls by measuring the distance to the nearest nest trees being used by adjacent pairs in the same year. We found no Barred Owl nest trees, so for this species we estimated min- imum nearest-neighbor distance by taking the mean of the 10 shortest distances between Barred Owl site-centers. We used Arc View version 3.1.1 (Environmental Systems Research Institute, Inc., Redlands, CA) to analyze spa- tially geographic, physiographic, vegetation, and owl data. GPNF geographic information system database sup- plied basic forest-age data. We divided the forested area into five age groups, as follows: 0-49-yr-old forest (mainly young trees growing on harvested areas); 50-79-yr-old forest (natural stands growing in burned areas without any late-successional characteristics); 80-129-yr-old forest (forest developing late-successional characteristics); 130- 179-yr-old forest (an equal sub-division between forest 280 yr old and forest <180 yr old to test for potential differences of younger and older mature forest) ; and for- est >180 yr old (forest with late-successional character- istics) . In western Washington, Douglas-fir stands usually begin to develop late-successional characteristics by the 268 Pearson and Livezey VoL. 37, No. 4 I ' ' 0 10 20 Km Figure 2. Locations of Spotted Owl occupied sites {N = 120, small dots), Spotted Owl unoccupied sites {N = 25, large dots). Barred Owl sites {N = 98, triangles), and potential Barred Owl sites {N = 76, boxes). Gray circles represent Spotted Owl home-range circles of 2.9-km radius. time they are ^80 yr old, including relatively large live and dead trees with some multi-layered canopies, and some large woody debris on the forest floor (USDA and USDI 1994a). We estimated the 1946 forest-age through- out the study area by subtracting 55 yr from the age in 2001 of each forested area. For areas where the 2001 forest-age was <55 yr, such as clearcut and burned areas, we estimated an original yr of origin based upon sur- rounding forest (e.g., a clearcut surrounded by forest with a 1729 yr-of-origin was given the same yr-of-origin) and then subtracted 55 yrs. We used nonparametric statistical tests because most of our data did not fit a normal distribution. Statistical tests (two-tailed Mann-Whitney Gtest, Spearman coefficient of rank correlation) were carried out using SYSTAT Version 10 (SPSS Inc., Chicago, IL). We followed Sokal and Rohlf (2000) to calculate tests. Significance level for all tests was a = 0.05. Sample sizes in statistical tests were pre- sented only when they departed from N = 500 for ran- dom sites, A = 145 for Spotted Owl sites and N = 98 for Barred Owl sites. All means were expressed as x ± 1 SD. For comparison of aspect between random. Spotted and Barred owl site-centers, we conducted chi-square tests of the frequency of site-centers within each of four directions (northeast, southeast, southwest, and north- west), north vs. south, and east vs. west, and we included only those random points that were in forest stands ^80 yr old {N — 250) to ensure that results would indicate possible selection rather than availability. Aspects for ran- dom site-centers were not equally distributed in each of the four directions due to naturally occurring differences in the landscape. We examined change in percentage of Barred Owl de- tections relative to all Strix detections. For this analysis, we determined the first and last years to be included in our sample of owl detections over time as follows: (1) we started with 1982 because it was the first yr with ^25 total Strix detections, and (2) we ended with 2000 because sur- December 2003 Spotted Owls and Barred Owls 269 Table 1. Mean (±SD) elevation (m) and slope (%) of Spotted Owl, Barred Owl, and random site-centers in the Cowlitz Valley Ranger District, Gifford Pinchot National Forest, Washington^ Spotted Owl Barred Owl Random {N= 145) (iV= 98) {N = 500) Elevation 966.2 ± 197.2’^-^ 812.5 ± 275.9^'d 1070.3 ± 282.6‘^’d Slope 54.1 ± 14.5'’’'^ 39.2 ± 18.6^'d 46.2 ± I7.16"'d ^ All significant Mann-Whitney P-values < 0.001. Mean differed from random site-centers. Mean differed from Barred Owl site-centers. Mean differed from Spotted Owl site-centers. veys in 2001 emphasized detecting Barred Owls more than in previous years. Because there were no Barred Owl detections in 1982, but there were Barred Owl de- tections in previous and in later years, we estimated the 1982 Barred Owl percentage as the mean of the detec- tions for this species for the 9 yr centered on 1982 (1978 — 86). We were unable to test the effect of Barred Owl presence on Spotted Owl site occupancy in sites that had not been harvested due to small sample sizes. Results Total Detections and Sites. We recorded 2170 Spotted Owl and 521 Barred Owl detections during July 1978-November 2001. Based on these detec- tions, we identified 145 Spotted Owl sites, 98 Barred Owl sites, and 76 potential Barred Owl sites in the study area (Fig. 2). Spotted Owl sites were centered on nests {N = 43), young {N = 41), lo- cations of adults of both sexes detected at the same time (N = 44) , locations of adults of both sexes detected at different times {N = 11), and resident singles {N = 6) . Barred Owl sites were centered on young (N = 8) , locations of adults of both sexes detected at the same time (N = 51), locations of adults of both sexes detected at different times {N = 21), and resident singles {N = 18). We believe we found virtually all of the Spotted Owl sites in the study area sometime during 1978-2001, and Figure 3. Percent of Barred Owl detections relative to all Strix detections by year, 1982-2000. that the actual total number of Barred Owl pair sites in our study area was probably closer to our total (A^ = 174) for both the Barred Owl sites and the potential Barred Owl sites. Distribution and Numbers. Both Spotted and Barred owl site-centers were significantly lower in elevation than random, and Spotted Owl site-cen- ters were significantly higher in elevation than Barred Owl site-centers (Table 1 ) . Spotted Owl site- centers were situated in areas of significantly steep- er slope than random and Barred Owl site-centers. Barred Owl site-centers were situated in areas of significantly less slope than random. None of the tests of aspect showed any significant difference be- tween Spotted Owl, Barred Owl, and random site- centers for the four directions (x^s = 1.63-5.99, P > 0.10) or for north vs. south or east vs. west (x^i = 0.30-3.30, P> 0.05). Table 2. Mean (±SD) area (ha) of five forest-age classes within 0,8-km radius plots (201 ha) around Spotted Owl, Barred Owl, and random site-centers in the Cowlitz Valley Ranger District, Gifford Pinchot National Forest, Wash- ington. Forest- age Class (yr) Spotted Owl {N= 145) Barred Owl (N = 98) Random {N = 500) 0-49 47.9 ± 32.1*^ 42.8 ± 33.0 43.5 ± 42.8^ 50-79 10.8 ± 26.3b 12.2 ± 33.6b 28.8 ± 52.4^’d 80-129 28.8 ± 42.2 32.3 ± 45.8 35.0 ± 51.2 130-179 17.2 ± 31.2b 14.2 ± 30.2 10.7 ± 24.7^^ >180 82.4 ± 44.6b 84.6 ± 54.1b 57.1 ± 49.6^=d Significant Mann-Whitney P-values ranged from P < 0.02 to P < 0.001. None of the means differed between Spotted Owl and Barred Owl sites (all Mann-Whitney P-values > 0.20). Mean differed from random core-plots. Mean differed from Spotted Owl core-plots. Mean differed from Barred Owl core-plots. 270 Pearson and Livezey VoL. 37, No. 4 Table 3. Age of forest stand at site-centers of Spotted Owl, Barred Owl, and random sites in the Cowlitz Valley Ranger District, Gifford Pinchot National Forest, Washington. Spotted Owt Barred Owl Random Forest-age Study Area {N = 145) (N = 98) (N = 442) a Class (yr) Ha % No. % No. % No. % 0-49 41951 24.1 0 0 1.0 Ill 25.1 50-79 28193 16.2 1 0.7 4 4.1 78 17.6 80-129 34 934 20.1 15 10.3 19 19.4 89 20.1 130-179 11814 6.8 10 6.9 7 7.1 26 5.9 >180 57 252 32.9 119 82.1 67 68.4 138 31.2 Total 174144 145 98 442b * The 58 random site-centers that fell in non-forested areas (rock, wet-mesic, dry meadow/brush) were excluded from this analysis. A 42-yr-old clearcut that retained some remnant old-growth trees. The percent of Barred Owl detections {N= 403) relative to all .Sfnx detections {N — 2431) increased significantly annually from 1982-2000 (t; = 0.790, P < 0.001, N= 19) . Overall, Barred Owl detections increased 8.6% annually from 1982-2000, and there did not appear to be any leveling-off of this increase (Fig. 3). Thirty-three percent (78 of 240) of the Strix detections in 2000 were Barred Owl detections, and 53% (113 of 215) of the Strix de- tections in 2001 (when detecting Barred Owls was emphasized) were Barred Owl detections. The minimum nearest-neighbor distances for Barred Owls and Spotted Owls were 1.1 km and 1.6 km, respectively. Hybrids. We detected two hybrids between Spot- ted and Barred owls. The first, which had a dis- tinctive five-note call, was detected once in 1994 and not afterward. The second hybrid, which had a distinctive six-note call, was audibly detected an- nually from 1995-2000. Site Characteristics. Spotted and Barred owl sites did not differ relative to the amount of different forest-age classes within core-plots (Table 2). On average, core-circles of both species contained more forest ^180 yr old and less forest 50-79 yr old than random plots. Spotted Owl plots also con- tained more 130-179-yr-old forest and 0-49-yr-old forest than random plots. Number of Spotted Owl site-centers in each for- est-age group differed from Barred Owls (x ^4 = 9.61, P < 0.05) and random (x ^4 = 130.31, P < 0.001), and that of Barred Owl site-centers differed from random (x ^4 = 61.56, P < 0.001; Table 3). Mean age of forest stand at site-centers was signif- icantly greater (U = 5844.5, P < 0.02) for Spotted Owls (254.7 ± 76.5 yr) than for Barred Owls (228.3 ± 101.5 yr). Thirty-three percent of forest in the study area was comprised of stands ^180 yr old, and 28% of random site-centers were in these old- forest stands. However, 82% of Spotted Owl and 68% of Barred Owl site-centers were in these old- forest stands (Table 3). Site Occupancy. Of the 129 Spotted Owl sites with ^10 surveys during the last 5 yr of the study, 25 (19.4%) apparently were unoccupied by Spot- ted Owls by 2001. There were significantly more Table 4. Mean (:tSD) elevation (m) and slope (%) of occupied and unoccupied Spotted Owl site-centers and Barred Owl site-centers in the Cowlitz Valley Ranger District, Gifford Pinchot National Forest, Washington. Occupied Unoccupied Spotted Owl Spotted Owi. Barred Owl {N= 104) {N= 25) {N= 98) Elevation 982.9 ± 143.9’’ 886.4 ± 291.8 812.5 ± 275.9^ Slope 57.1 ± 12.P’-’ 43.0 ± 19.2^’ 39.2 ± 18.6d ^ All significant Mann-Whitney T-values < 0.001. Mean differed from Barred Owl site-centers. Mean differed from unoccupied Spotted Owl site-centers. Mean differed from occupied Spotted Owl site-centers. December 2003 Spotted Owls and Barred Owls 271 AWA CHU MAT AMA LSR CRA Land Use Allocation Figure 4. The percent of Spotted Owl sites that were occupied (circles) compared to the percent of Barred Owl sites relative to all occupied Strix sites (triangles) by Land Use Allocation. Timber harvest was permitted in critical habitat units (CHU), matrix (MAT), and the adaptive management area (AMA), and was not permit- ted in administratively-withdrawn areas (AWA), late-suc- cessional reserves (LSR) , and congressionally-reserved ar- eas (CRA). Barred Owl site-centers in: (1) unoccupied (0.44 ± 0.51) than occupied (0.14 ± 0.35) Spotted Owl cir- cles of 0.8-km radius {U = 915.5, P = 0.001), (2) unoccupied (0.84 ± 0.99) than occupied (0.44 ± 0.69) Spotted Owl circles of 1.6-km radius {U = 1016.0, P = 0.049), and (3) unoccupied (2.40 ± 1.83) than occupied (1.32 ± 1.38) Spotted Owl home-range circles of 2.9-km radius (t/ = 845.5, P = 0.005). Occupied Spotted Owl sites were significantly steeper in slope and were significantly higher in elevation than Barred Owl sites (Table 4). Com- pared to unoccupied Spotted Owl sites, occupied Spotted Owl sites were also significantly steeper in slope, but were not significantly different in ele- vation. Unoccupied Spotted Owl sites were not sig- nificantly different than Barred Owl sites in slope or elevation. The number of ha of forest >80 yr old was not significantly different in occupied (130.1 ± 36.3 ha) vs. unoccupied (123.4 ± 37.8 ha) (U= 1445.0, P — 0.388) Spotted Owl core-plots. However, when considering core-plots in which some forest harvest had taken place from 1978-2001 (20.6 ± 16.2 ha, range = 0.1—72.6 ha), occupied sites (125.5 ± 31.4 ha, N = 83) had significantly more forest ^80 yr old than unoccupied sites (109.9 ± 28.6 ha, N = 19) (U= 1022.0, P = 0.045). The percent of Spotted Owl sites that were oc- cupied in each LUA was inversely proportional to the percent of Barred Owl sites relative to all oc- cupied Strix sites (r^ = —0.899, P < 0.01, A — 6; Table 5. Reserve areas and timber harvest relative to number of occupied Spotted Owl sites {N = 104), Barred Owl sites, and all Barred Owl (including potential Barred Owl sites) sites {N = 174) in the Cowlitz Valley Ranger District, Gifford Pinchot National Forest, Washington. Timber Management Late Har- Crit- ICjVL Habitat SUCCES- SIONAL Reserve No Har- vest^ VEST Per- mitted' Spotted Owl sites 57 35 44 60 Barred Owl sites 55 47 64 34 All Barred Owl sites 89 73 102 72 ^ Includes late-successional reserve, congressionally-reserved are- as, and administratively-withdrawn areas. Includes matrix and adaptive management area. Fig. 4). In areas specifically allocated to benefit Spotted Owls (critical habitat units and late-succes- sional reserves), the number of Barred Owl sites approached or surpassed the number of occupied Spotted Owl sites (Table 5). The numbers of oc- cupied Spotted Owl sites and Barred Owl sites dif- fered between areas in which timber harvest was permitted and was not permitted (x^i = 10.77, P < 0.005). There were more Spotted Owl sites than Barred Owl sites in areas with timber harvest and fewer Spotted Owl sites than Barred Owl sites in areas without timber harvest. When including po- tential Barred Owl sites, there were more than twice as many Barred Owl sites than Spotted Owl sites in areas without timber harvest, whereas in areas with timber harvest, there were slightly more Barred Owl sites than Spotted Owl sites (Table 5) . Discussion Distribution and Numbers. Within the study area, more gradual slopes were along valley-bot- toms and, in some cases, higher-elevation plateaus. Barred Owls generally appeared to favor the less- steep valley bottoms, while Spotted Owls were found more often in the steeper upland areas, a distinction also noted by Herter and Hicks (2000) . Our analysis did not show that aspect had any in- fluence on site selection by Spotted or Barred owls. Although studies in comparatively warmer regions of California indicated that Spotted Owls may se- lect roosts or nests on north aspects during sum- mer (Barrows 1981, North et al. 2000), studies in the more mesic conditions typical of western 272 Pearson and Livezey VoL. 37, No. 4 Table 6. Mean (±SD) area (ha) of five forest-age classes in 0.8-km radius plots (201 ha) around 1946 Spotted Owl, 2001 Spotted Owl, 1946 random, and 2001 random site-centers in the Cowlitz Valley Ranger District, Gifford Pinchot National Forest, Washington. Forest-age Class (yrs) 1946 Spotted Owl {N= 145) 2001 Spotled Owt (V= 145) 1946 Random {N = 500) 2001 Random {N= 500) 0-49 22.4 ± 39.3‘=>d>^ 47.9 ± 32.U" 46.8 ± 63.9^.'= 43.5 ± 42.8b.'=.d 50-79 18.9 ± 38.2 10.8 ± 26.3'= 21.7 ± 45.9^ 8.8 ± 52.4">'i 80-129 17.5 ± 33.0"-^ 28.8 ± 42.4»^ 10.3 ± 24.5*’'^ 35.0 ± 51.2'^ 130-179 22.0 ± 43.5 17.2 ± 31.2^ 16.2 ± 34.6 10.7 ± 24.7^= >180 106.1 ± 59.6"'d 82.4 ± 44.6^-'= 80.2 ± 70.5*^ 57.1 ± 49.6^’'i ® Significant Mann-Whitney P-values ranged from P < 0.020 to P < 0.001. Mean differed from 1946 Spotted Owl core-plots. Mean differed from 2001 Spotted Owl core-plots. Mean differed from 1946 random core-plots. Mean differed from 2001 random core-plots. Oregon and Washington indicated little selection for aspect by Spotted Owls (Forsman et al. 1984, this study). At the northern edge of the range of the Spotted Owl, surveys suggested that there were four times as many Barred Owl sites than Spotted Owl sites both in British Columbia during the late 1980s (Dunbar et al. 1991) and in North Cascades Na- tional Park, Washington, during 1993-96 (R. Kuntz and R. Christopherson unpubl. data) . Hamer et al. (1989) found that Barred Owls were twice as abun- dant as Spotted Owls in the northern Cascade Mountains east of Mt. Baker in the late 1980s. Barred Owls were almost as numerous as Spotted Owls in the mid 1990s just north of Mt. Rainier (Herter and Hicks 2000) and, in our study area, it appeared that they were at least as numerous as Spotted Owls in 2001. Our shorter minimum nearest-neighbor distance for Barred Owls vs. Spotted Owls, which also was found by Hamer et al. (1989), could be a result of the wider variety of prey that Barred Owls use (Ha- mer et al. 2001), thereby allowing Barred Owls to “pack” closer together. The shortest nearest-neigh- bor distance for Spotted Owls (1.6 km) in Forsman et al. (1984) was equal to our shortest nearest- neighbor distance. Hybrids. Apparently, Spotted X Barred owl hy- brids are rare. Kelly (2001) gathered reports of vi- sual sightings of only 24 adult and 26 juvenile Spot- ted X Barred owl hybrids in Washington and Oregon from 1974-99. Site Characteristics. Our data suggest that Spot- ted Owl site-centers were more often in older for- est than Barred Owl and random site-centers. Also, territories of Spotted and Barred owls tended to be centered in areas characterized by higher con- centrations of old forest and lower concentrations of 50-79-yr-old forest than were available at ran- dom. These findings were in agreement with most previous studies of Spotted Owls (Bart and Fors- man 1992, Hunter et al. 1995, Swindle et al. 1999), but the similarity of forest age in Spotted and Barred owl core-plots was a somewhat surprising result. Barred Owls have been thought to be more habitat generalists than Spotted Owls (Hamer et al. 1989, Herter and Hicks 2000, Hamer et al. 2001), and seem to use much more varied habitats throughout the Pacific Northwest (Smith et al. 1997, Kelly and Forsman 2003). However, in the eastern United States and eastern Canada, Barred Owls are considered an old-growth forest species (Dunstan and Sample 1972, Devereux and Mosher 1984, Elody and Sloan 1985, Laidig and Dobkin 1995) and, in some areas. Barred Owls do not breed in or defend territories in younger forests (Haney 1997). In Manitoba, Barred Owl plots con- tained fewer clearcut, burned, and young forest ar- eas than random plots, and were associated more with areas of high crown-closure than random plots (Hinam and Duncan 2002). Because Barred Owls use forests that are similar to those used by Spotted Owls in our study area, this may make it difficult to manage forests to benefit Spotted Owls over Barred Owls. Spotted Owl core-plots in our study area con- tained more old forest (^130 yr) than random sites, but contained less forest 50-79 yr old than Spotted Owls and Barred Owls 273 December 2003 0 10 20 km Figure 5. Forest s80 yr old in 1946 that was lost to forest harvest by 2001 (black areas), forest >80 yr old that remained in 2001 (dark gray areas), forest that became 80 yr old from 1946-2001 (light gray areas), and the 145 Spotted Owl site-centers (circles). random. This finding, coupled with results from other studies that Spotted Owls are an old-forest associated species (USDA and USDI 1994a, 1994b), could lead to an expectation that Spotted Owl core-plots should contain less forest 0-49 yr old than random areas. However, Spotted Owl core- plots in our study area contained significantly more ha of 0-49-yr-old forest than random plots. To further examine this counter-intuitive result, we adjusted stand age to approximate the forest in the study area before any forest harvest took place (in 1946) to compare the forest-age distribution in Spotted Owl and random sites in 1946 vs. 2001 (Ta- ble 6). Spotted Owl core-plots in 1946 contained significantly less 0-49-yr-old forest than 1946 ran- dom sites, with a mean less than one-half that of random (22.8 vs. 46.8 ha) . In 1946, the 0-49-yr-old forest was consolidated into large, contiguous areas caused by forest fires in the early 1900s, while in 2001, 0-49-yr-old forest, the result of thousands of clearcuts, was widely distributed and was inter- mixed with older forest. The distribution of Spot- ted Owl sites in 2001 mirrored very closely the dis- tribution of forest ^80 yr old in 1946, excluding the areas lost to forest harvest. Spotted Owls ap- peared to have persisted in areas of older forest. Other than a few instances. Spotted Owls had yet to repopulate the areas that had grown into forest ^80 yr old by 2001 (Fig. 5). Spotted Owl core-plots in our study area includ- ed more forest ^130 yr old than random, but did not contain more forest 80-129 yr old than ran- 274 Pearson and Livezey VoL. 37, No. 4 dom. However, forest practices reduced the amount of forest >130 yr old by 25 921 ha and increased the amount of forest 80-129 yr old by 22 672 ha. Replacing stands Si 30 yr old with stands 80-129 yr old could negatively affect Spot- ted Owls. Our results were similar to previous studies in that Barred Owls were most abundant in lowland forests and less common in upland areas (Laidig and Dobkin 1995, Haney 1997, Mazur et al. 1997, Hamer 1988, Mazur et al. 1998). However, these owls also occurred in upland areas, as has been reported in previous studies (Boxall and Stepney 1982, Dunbar et al. 1991, Dark et al. 1998, Wright and Hayward 1998). Collectively, these results sug- gest that riparian zones and lowland forests were more suitable habitat for Barred Owls in our study area, and that upland forests were less likely to be occupied by Barred Owls. The large standard deviations for elevation and slope (Table 1) indicated considerable variation among sites occupied by both Barred Owls and Spotted Owls. For example, there were four cases where Spotted Owl core-plots were dominated by 50-79-yr-old forest (53-96% cover), with only small areas of old forest or remnant older trees remain- ing. Site Occupancy. When we analyzed only those core-plots in which some timber harvest had taken place since 1978, unoccupied Spotted Owl sites had significantly less forest ^80 yr old remaining than occupied sites, indicating that forest harvest contributed to these sites becoming unoccupied. Occupancy of Spotted Owl sites, relative to (1) number of Barred Owl sites within Spotted Owl plots, (2) slope of Spotted Owl sites, and (3) ele- vation of Spotted Owl sites could be directly or in- directly traced to the presence of Barred Owls. There were significantly more Barred Owl sites within 0.8-km, 1.6-km, and 2.9-km radius circles centered on Spotted Owl site-centers in unoccu- pied Spotted Owl sites than in occupied Spotted Owl sites. Occupied Spotted Owl sites were on sig- nificantly steeper slopes and were significantly higher in elevation than Barred Owl sites, whereas unoccupied Spotted Owl sites were not significant- ly different than Barred Owl sites in slope or ele- vation. Our results suggest that Spotted Owls are more likely to abandon a site if Barred Owls take up res- idence close to that site. Similarly, Kelly et al. (2003) found that occupancy of Spotted Owl sites declined after Barred Owls were detected within 0.8 km, but occupancy was not affected when Barred Owls were located >0.8 km from Spotted Owl site-centers. We suggest that a combination of habitat lost due to timber harvest and the presence of Barred Owls may work synergistically to put Spotted Owl pairs at risk of losing their territories. Sites in the eastern Washington Cascade Moun- tains, unoccupied for 4 (N = 2), 6 (N = 3), and 8 (N = 1) yr were reoccupied by Spotted Owls; the marked, replacement Spotted Owls in these cases tended to use the same core areas and, in some sites, even nested in the same nest tree used by previous owls (T. Fleming pers. comm.). There- fore, some of our unoccupied sites may become reoccupied in the future. Spotted Owl Conservation. Existing metapopu- lation conservation strategies of the northern Spot- ted Owl are dependent upon Spotted Owls surviv- ing and reproducing in reserves, and being able to move between reserves via hospitable habitats to facilitate genetic interchange and, if necessary, re- colonization (Levins 1968, USDA and USDI 1994a, Caughley and Gunn 1996, Gutierrez and Harrison 1996, Noon and McKelvey 1996). Our data sug- gested that Barred Owls caused a reduction in the Spotted Owl population by physically excluding them from historic territories and making those territories unavailable for recolonization, as was suggested by Dunbar et al. (1991), Dark et al. (1998), and Kelly (2001). There is an increasing body of evidence that Barred Owls physically attack Spotted Owls (E. Forsman, J. Mowdy, T. Snetsinger, and G. Stagner pers. comm.), and sometimes may kill them (Leskiw and Gutierrez 1998). Aside from direct competition for space, it is also likely that Barred and Spotted owls compete for prey (Hamer et al. 2001). In addition, Barred Owls may nega- tively affect dispersing, juvenile Spotted Owls by creating a hostile environment that inhibits the oc- cupation of vacated Spotted Owl territories and other suitable areas. In our study area, presence of Barred Owls had a greater effect on Spotted Owl site occupancy than did the status of the area as a Spotted Owl reserve. In fact, our data suggest that Barred Owls were more numerous in the reserve areas than Spotted Owls. Continued loss of old- growth and mature forest may reduce the ability of Spotted Owls to persist in the presence of Barred Owls. We recommend continuing the long-term Spot- ted Owl demography studies (Franklin et al. 1999) December 2003 Spotted Owi.s and Barred Owls 275 to track occupancy and reproduction of northern Spotted Owls range-wide. Also, this work should in- clude surveys for Barred Owls to determine if the presence of this species affects Spotted Owl occu- pancy and reproduction. Acknowledgments Adena Carson provided support for surveys. U.S. For- est Service employees Tom Kogut and Joe Kulig supplied valuable insights regarding owls within the study area. Tracy Fleming provided helpful information and review. We are grateful for comments on various drafts by Joe Buchanan, Rocky Gutierrez, Karl Halupka, Kit Hershey, Dale Herter, Elizabeth Kelly, Bradley Livezey, Michael Morrison, Pam Repp, Sarah Rinkevich, David Solis, and Keith Swindle. Review comments by Eric Forsman were especially helpful. Pam Repp encouraged the writing of this paper, Bradley Livezey and John Sauer provided sta- tistical advice, Jim Petterson supplied locations of Spotted Owl sites in Mt. Rainier National Park used for nearest- neighbor analyses, and Nancy Magh helped translate the resumen. We are grateful for the use of U.S. Forest Service data. Robert Pearson’s fieldwork was partially supported by the U.S. Forest Service (1991-92), Washington State Department of Natural Resources (1993), and the U.S. Fish and Wildlife Service (2001). Literature Cited Anthony, R.G. 2001. Low productivity of Bald Eagles in Prince of Wales Island, southeast Alaska. /. Raptor Res. 35:1-8. Barrows, C.C. 1981. Roost selection by Spotted Owls: an adaptation to heat stress. Condor 83:302—309. Bart, J. and E.D. Forsman. 1992. Dependence of north- ern Spotted Owls Strix occidentalis caurina on old- growth forests in the western USA. Biol. Conserv. 62: 95-100. Boxall, P.C. and P.H.R. Stepney. 1982. The distribution and status of the Barred Owl in Alberta. Can. Field- Nat. 96:46-50. Gaughley, G. and a. Gunn. 1996. Conservation biology in theory and practice. Blackwell Science, Inc., Cam- bridge, MA U.S.A. Dark, S.J., R.J. Gutierrez, and G.I. Gould, Jr. 1998. The Barred Owl {Strix varia) invasion in California. Auk 115:50-56. Devereaux, J.G. and J.A. Mosher. 1984. Breeding ecol- ogy of Barred Owls in the central Appalachians. Rap- tor Res. 18:49-58. Dunbar, D.L., B.P. Booth, E.D. Forsman, A.E. Hethe- rington, and D.J. Wilson. 1991. Status of the Spotted Owl, Strix occidentalis, and Barred Owl, Strix varia, in southwestern British Columbia. Can. Field-Nat. 105: 464-468. Dunstan, T.C. and S.D. Sample. 1972. Biology of Barred Owls in Minnesota. Loon 44:111-115. Elody, B.I. and N.F. Sloan. 1985. Movements and habitat use of Barred Owls in the Huron Mountains of Mar- quette County, Michigan, as determined by radiote- lemetry. Jack-Pine Warbler 63:3-8. Evens, J. and R. LeValley. 1982. The spring migration, middle Pacific coast region. Am. Birds 36:889-892 Forsman, E.D. 1983. Methods and materials for locating and studying Spotted Owl. U.S. Dept, of Agriculture Forest Service Gen. Tech. Rpt. PNW-162. Pacific Northwest Forest and Range Experiment Station, Cor- vallis, OR U.S.A. , E.C. Meslow, and H.M. Wight. 1984. Distribu- tion and biology of the Spotted Owl in Oregon. Wildl Monogr, 87. Frankijn, A.B., K.P. Burnham, G.C. White, R.G. Ant- hony, E.D. Forsman, C. Schwarz, J.D. Nicols, andJ. Hines. 1999. Range-wide status and trends in north- ern Spotted Owl populations. Colorado Coop. Fish and Wildl. Res. Unit, Fort Collins, Colorado and Oregon Coop. Fish and Wildl. Res. Unit, Corvallis, OR U.S.A. Gutierrez, RJ. 1994. Changes in the distribution and abundance of Spotted Owls during the past century. Stud. Avian Biol. 15:293-300. and S. Harrison. 1996. Applying metapopulation theory to Spotted Owl management; a history and cri- tique. Pages 167-185 in D.R. McCullough [Ed.], Me- tapopulations and wildlife conservation. Island Press, Covelo, CA U.S.A. Hamer, T.E. 1988. Home range size of the northern Barred Owl and northern Spotted Owl in western Washington. M.S. thesis. Western Washington Univer- sity, Bellingham, WA U.S.A. , S.G. Seim, and K.R. Dixon. 1989. Northern Spot- ted Owl and northern Barred Owl habitat use and home range size in Washington: preliminary report. Washington Department of Wildlife, Olympia, WA U.S.A. , E.D. Forsman, A.D. Fuchs, and M.L. Walters. 1994. Hybridization between Barred and Spotted owls. 111:487-492. , D.L. Hays, C.M. Senger, and E.D. Forsman. 2001. Diets of northern Barred Owls and northern Spotted Owls in an area of sympatry. /. Raptor Res. 35. 221-227. Haney, J.C. 1997. Spatial incidence of Barred Owl {Stnx varia) reproduction in old-growth forest of the Ap- palachian Plateau./. Raptor Res. 31:241-252. Herter, D.R. and L.L. Hicks. 2000. Barred Owl and Spot- ted Owl populations and habitat in the central Cas- cade Range of Washington. /. Raptor Res. 34:279-286. Hinam, H.L. and J.R. Duncan. 2002. Effects of habitat fragmentation and slope on the distribution of three owl species in the Manitoba escarpment, Canada — a preliminary analysis. Pages 148-161 in I. Newton, R Kavanaugh, J. Olsen, and I. Taylor [Eds.], Ecology and conservation of owls. CSIRO Pubk, Collingwood, Australia. Hunter, J.E., RJ. Gutierrez, and A.B. Franklin. 1995. 276 Pearson and Livezey VoL. 37, No. 4 Habitat configuration around Spotted Owl sites in northwestern California. Condor 97:684— 693. Kelly, E.G. 2001. Range expansion of the northern Barred Owl: an evaluation of the impact on Spotted Owls. M.S. thesis, Oregon State University, Corvallis, OR U.S.A. AND E.D. Forsman. 2003. Barred Owl Pages 320- 322 in D.B. Marshall, M.G. Hunter, and L.L. Contre- ras [Ed.], Birds of Oregon, a general reference. Oregon State University, Corvallis, OR U.S.A, , E.D. Forsman, and R.G. Anthony. 2003. Are Barred Owls replacing Spotted Owls? Condor 105:45- 53. Laidig, K.J. and D.S. Dobkin. 1995. Spatial overlap and habitat association of Barred Owls and Great Horned Owls in southern Newjersey./. Raptor Res. 29:151—157. Leskiw, T. and R.J. Gutierrez. 1998. Possible predation of a Spotted Owl by a Barred Owl. West. Birds 29:225— 226. Uevins, R. 1968. Evolution in changing environments: some theoretical explorations. Monographs in popu- lation biology, Princeton Univ. Press, Princeton, NJ U.S.A. Mazur, K.M., P.C. James, MJ. Fitzsimmons, G. Uangen, AND R.H.M. Espie. 1997. Habitat associations of the Barred Owl in the boreal forest of Saskatchewan, Ca- nada, y. Raptor Res. 31:253-259. , S.D. Frith, and P.C. James. 1998. Barred Owl home range and habitat selection in the boreal forest of central Saskatchewan. Auk 115:746-754. McGarigal, K. and J.D. Fraser. 1985. Barred Owl re- sponses to recorded vocalizations. Condor 87:552-553. Meyer, J.S., U.U. Irwin, and M.S. Boyce. 1998. Influence of habitat abundance and fragmentation on northern Spotted Owls in western Oregon. Wildl. Monogr. 139. Noon, B.R. and K.S. McKelvey. 1996. A common frame- work for conservation planning: linking individual and metapopulation models. Pages 139-165 in D.R. McCullough [Ed.], Metapopulations and wildlife con- servation. Island Press, Covelo, CA U.S.A. North, M., G. Steger, R. Denton, G. Eberlein, T. Mun- TON, and K. Johnson. 2000. Association of weather and nest-site structure with reproductive success in California Spotted Owls./. Wildl. Manage. 64:797-807. Reid, J.A., R.B. Horn, and E.D. Forsman. 1999. Detec- tion rates of Spotted Owls based on acoustic-lure and live-lure surveys. Wildl. Soc. Bull. 27:986-990. Rignaix, J.A. 1973. Owls of the world. E.P. Dutton & Co., Inc., New York, NY U.S.A. Rogers, T.H. 1966. The fall migration. Northern Rocky Mountain-intermountain region. Aud. Field Notes 20: 74. Smith, M.R., P.W. Mattocks, Jr., and K.M. Cassidy. 1997. Breeding birds of Washington state, Vol. 4. In K.M. Cassidy, C.E. Grue, M.R, Smith, and K.S. Dvornich, [Eds.], Washington State Gap Analysis: Final Report, Seattle Audubon Society Pubk, Seattle, WA U.S.A. Sokal, R.R. and EJ. Rohlf. 2000. Biometry. W.H. Free- man and Co., New York, NY U.S.A. Swindle, K.A., WJ. Ripple, E.C. Meslow, and D. Scha- fer. 1999. Old-forest distribution around Spotted Owl nests in the central Cascade Mountains, Oregon. J. Wildl. Manage. 63:1212-1221. Taylor, A.U. and E.D. Forsman. 1976. Recent range ex- tensions of the Barred Owl in western North America, including the first records for Oregon. Condor 78:560- 561. U.S. Department of Agriculture and U.S. Department OF the Interior. 1994a. Record of decision for amendments to Forest Service and Bureau of Uand Management planning documents within the range of the northern Spotted Owl, and Standards and Guide- lines for management of habitat for late-successional and old-growth forest related species within the range of the northern Spotted Owl. U.S. Dept, of Agricul- ture Forest Service and U.S. Dept, of the Interior Bu- reau of Land Management, Portland, OR U.S.A. AND . 1994b. Final supplemental environ- mental impact statement on management of habitat for late-successional and old-growth forest related spe- cies within the range of the Spotted Owl. U.S. Dept, of Agriculture Forest Service and U.S. Dept, of the Interior Bureau of Land Management, Portland, OR U.S.A. Wright, A.I.. and G.D. Hayward. 1998. Barred Owl range expansion into the central Idaho wilderness./. Raptor Res. 32:77-81. Received 17 June 2002; accepted 29 July 2003 Associate Editor: Clint W. Boal J. Raptor Res. 37(4):277-285 © 2003 The Raptor Research Foundation, Inc. TEMPORAL AND SPATIAL STABILITY OF RED-TAILED HAWK TERRITORIES IN THE LUQUILLO EXPERIMENTAL FOREST, PUERTO RICO Clint W. Boal^ USGS-BRD Texas Cooperative Fish and Wildlife Research Unit, Texas Tech. University, Lubbock, TX 79409-2120 US. A. Helen A. Snyder RO. Box 16426, Portal, AZ 85632 US. A. Brent D. Bibles Utah State University-Uintah Basin, 1680 W. Hwy. 40, Vernal, UT 84078 US. A. Tracy S. Estabrook Department Range, Wildlife and Fisheries Management, Texas Tech. University, Lubbock, TX 79409 US. A. Abstract. — ^We mapped Red-tailed Hawk {Buteo jamaicensis) territories in the Luquillo Experimental Forest (LEF) of Puerto Rico in 1998. We combined our 1998 data with that collected during previous studies of Red-tailed Hawks in the LEF to examine population numbers and spatial stability of territorial boundaries over a 26-yr period. We also investigated potential relationships between Red-tailed Hawk territory sizes and topographic and climatic factors. Mean size of 16 defended territories during 1998 was 124.3 ± 12.0 ha, which was not significantly different from our calculations of mean territory sizes derived from data collected in 1974 and 1984. Aspect and slope influenced territory size with the smallest territories having high slope and easterly aspects. Territory size was small compared to that reported for other parts of the species’ range. In addition, there was remarkably little temporal change in the spatial distribution, area, and boundaries of Red-tailed Hawk territories among the study periods. Further, there was substantial boundary overlap (21-27%) between defended territories among the different study periods. The temporal stability of the spatial distribution of Red-tailed Hawk territories in the study area leads us to believe the area might be at or near saturation. Key Words: Buteo jamaicensis; Red-tailed Hawk, Puerto Rico', spatial stability, temporal stability, territorial characteristics. ESTABILIDAD TEMPORAL Y ESPACIAL DE LOS TERRITORIOS DEL GAVILAN DE COLA ROJA EN EL BOSQUE EXPERIMENTAL DE LUQUILLO, PUERTO RICO. Resumen. — Hicimos mapas de los territorios del gavilan de cola roja {Buteo jamaicensis) en el Bosque Experimental de Luquillo (BEL) en Puerto Rico en 1998. Combinamos nuestros datos de 1998 con aquellos colectados durante estudios previos con gavilanes de cola roja en el BEL para examinar los numeros de la poblacion y la estabilidad espacial de los limites de los territorios en un periodo de 26 anos. Ademas investigamos las posibles relaciones entre los tamafios de los territorios del gavilan y factores topograficos y climaticos. El tamano medio de los 16 territorios defendidos durante 1998 fue 124.3 ± 12.0 ha, lo cual no fue significativamente diferente de nuestros calculos del tamano medio de los territorios a partir de datos colectados en 1974 y 1984. La cara y la pendiente influyeron en el tamano del territorio, teniendo los territorios mas pequenos altas pendientes y caras orientales. El tamano del territorio fue pequeno comparado con los reportados para otras partes dentro del rango de las especie. Adicionalmente, hubo notablemente poco cambio temporal en la distribucion espacial, area y limites de los territorios del gavilan de cola roja entre loas periodos estudiados. Ademas hubo un traslape substancial en los limites (21-27%) en los territorios defendidos entre los diferentes perio- ^ E-mail address: cboal@ttacs.ttu.edu 277 278 Boal et al. VoL. 37, No. 4 dos estudiados. La estabilidad temporal en la distribucion espacial de los territorios del gavilan de cola roja en el area de estudio no lleva a creer que el area puede estar en o cerca de su saturacion. [Traduccion de Cesar Marquez] The Red-tailed Hawk {Buteo jamaicensis) is pos- sibly the most ubiquitous raptor in North America, with a distribution from Alaska and northern Ca- nada south through most of Mexico and into Cen- tral America, and east across the West Indies (Sny- der and Snyder 1991). It occupies a diversity of ecotypes, but is typically considered a raptor of open country characterized by interspersed wood- lots (Preston and Beane 1993). Red-tailed Hawks are primarily sit-and-wait predators, scanning open areas from elevated perch sites (Preston and Beane 1993). However, these hawks also occupy tropical forest areas (Snyder and Snyder 1991, Preston and Beane 1993). The Red-tailed Hawk population in the Luquillo Experimental Forest (LEE) of Puerto Rico is one of the densest (1.6 pairs/km^) ever re- ported for the species (Snyder and Snyder 1991, Preston and Beane 1993). Furthermore, Santana and Temple (1988) found the density of Red-tailed Hawks was greatest in the mountainous rain forest of the LEE and lowest in the open-country low- lands. This is intriguing because the LEE is a closed-canopy forest and quite atypical from the usual open and mixed vegetation ecotypes com- monly occupied by Red-tailed Hawks (Howell et al. 1978, Preston and Beane 1993). We mapped Red-tailed Hawk territories in the LEE study area in 1998. Here we combine our data with that collected during previous studies of these species in the LEE from 1972-75 (Snyder et al. 1987; hereafter 1970s study) and from 1981-83 (Santana and Temple 1988; hereafter 1980s study) to examine population stability and the spatial sta- bility of territorial boundaries across three de- cades. We also explore potential relationships of topographic and climatic factors to territory sizes in the study area. Si UDY Area Our study area was an irregularly shaped 32-km^ area within the Luquillo Experimental Forest (Snyder et al. 1987). The LEF is a subtropical rain forest in the Luqui- llo Mountains, located at the east end of the island of Puerto Rico (18°N, 65°W), and is part of the Caribbean National Eorest in the U.S. Department of Agriculture National Forest system. The study area ranges from 300- 1050 m in elevation and includes “tabonuco forest” (sub- tropical wet forest life zone), “Colorado forest” (lower montane wet forest life zone), and “cloud forest” (lower montane rain forest life zone) (Ewel and Whitmore 1973). In general, as elevation increases, mean tree height, diameter, basal area, and species richness decreas- es, while stem density increases (Brown et al. 1983). Annually, the study area receives more than 1000 rain showers, with rainfall ranging from 354-485 cm (Garcia- Martino et al. 1996). Temperatures at the highest eleva- tions range from 17-20°C (Garcia-Martino et al. 1996). Methods A territory is usually considered as that area in which a resident individual or pair excludes conspecifics (Pow- ell 2000). Red-tailed Hawks in the LEF defend their ter- ritories year-round, attacking any conspecifics that stray into their territory (Snyder et al. 1987). Observations of intraspecific encounters between neighboring Red-tailed Hawks allow mapping of territory boundaries (Janes 1984, Santana and Temple 1984, Snyder et al. 1987). Intraspecific territorial encounters typically involve tbe resident Red-tailed Hawk diving from a soar or initiating a direct flapping chase. Intruders are pursued by one or both members of the resident pair to a boundary of the territory, at which point the intruder stops fleeing and both resident and intruder will circle up near each other while vocalizing. The interaction ends when both resi- dent and intruder retreat into their respective territories, often with one individual of a pair flying above and to the rear of its mate, after which the pairs drop down into their territories, presumably near the nest area. The se- quence of soaring residents chasing intruding neighbors to a boundary, then circling and screaming until both sets of hawks return to the relative centers of their ter- ritories may be repeated every few hours throughout the day with several neighboring conspecifics. The chases consistently stop at the same locations, which we inter- preted as the territorial boundaries. The hunting behaviors of Red-tailed Hawks in the LEF also facilitate territory mapping. The almost constant up- drafts, created by a northeasterly laminar flow of air against the Luquillo Mountains, allow Red-tailed Hawks to expend little energy while hunting in flight (Snyder and Snyder 1991). Thus, the majority of hunting occurs while in flight rather than from a perch (Snyder et al. 1987, Santana and Temple 1988). We replicated earlier studies of Red-tailed Hawks in the LEF (Snyder et al. 1987, Santana and Temple 1988) by making visual observations of hunting Red-tailed Hawks and their interactions with conspecific neighbors and in- truders from 5-20 January 1998, which is the onset of the nesting period in the study area (Snyder et al. 1987). We made our observations from many of the same over- looks used by Snyder et al. (1987) and Santana and Tem- ple (1988). These consisted primarily of clear-viewing lo- cations along roads, the top of rocks, slides, hillsides and escarpments, and the Mt. Britton and Yokahu lookout towers. Some of the previously-used road overlooks were no longer suitable due to vegetation changes. Fog pre- cluded use of some of the high-elevation sites on some days, so we made the sampling from these lookouts a December 2003 Red-tailed Hawk Territories in Puerto Rico 279 priority when good viewing conditions prevailed. We be- gan observations at about 0800 H and continued until 1700 H unless fog or rain developed. We recorded data directly onto photocopies of a 1981 1:20 000 U.S. Forest Service topographic map with 10-m contour intervals. We used the map copies to record observed flight lines for individual hawks, locations of perch points, copulation sites, nest sites, and locations of intraspecihc interactions relative to obvious landmarks. We numbered observa- tions sequentially and recorded observation times, vocal- izations, details of individual birds’ molts and markings, and the details of intraspecific interactions observed. We used fresh map copies each day and used additional map photocopies during the day as needed to maintain clarity of the records. Due to the amount of time they spent in flight, hawks were easiest to detect initially when soaring aloft or when sunlight reflected off their wings as they circled. Hawks flying against the background of forest canopy were hard- er to locate and track visually. When we identified regu- larly-used perches, we subsequently could often locate perched hawks and observe them as they moved within their defended area. Once an individual hawk was sight- ed, we visually tracked the bird until it was lost from sight in clouds, the forest canopy, or behind topographical fea- tures. Hawks were in view for periods ranging from sev- eral minutes to an hour or more. When interactions be- tween two or more Red-tailed Hawks occurred, each observer chose an individual or a pair to follow exclu- sively. Thus, it was advantageous to have multiple observ- ers. We easily plotted locations of perched hawks and points where hawks dove under the canopy. Apart from brief and intermittent cloud buildup and subsequent rain showers, sunlight was always available and we esti- mated locations of hawks flying within ca. 10 m of the forest canopy using the shadow they cast on the canopy. Errors of a few meters due to sun angle likely had only minor influence on our estimation of hawk locations. Ob- servations were facilitated by the topography of the study area; we could accurately gauge a hawk’s position and movements relative to cliff faces, drainage bottoms, and other topographic or landmark features of known size, elevation, and distance from the observation point. Fi- nally, two observers frequently recorded data indepen- dently from different observation points. The resulting estimations of hawk locations were then compared be- tween the two observers to verify the location plotted and improve the accuracy. In these ways, we were able to de- velop a sense of hawk size at different distances and make estimations of their location relative to the ground when they were soaring. Although some error undoubtedly oc- curred, we used methods identical to those used during the 1970s (Snyder et al. 1987) and 1980s (Santana and Temple 1988), so estimates should have equivalent bias and comparisons among study periods are valid. Based on the fact that we estimated visual locations against a topographically complex background at distances of only a few hundred meters, we suggest that our error is not substantially different from, and may be less than, error associated with radiotelemetry triangulations. We used location points, flight lines, and behavioral observations to map the estimated area of each pair’s territory. We used ArcView 3.2 (Environmental Systems Research Institute, Inc.) to plot outermost location points for members of each Red-tailed Hawk pair onto a digital map of the study area. We then employed the ArcView Animal Movements extension (Hooge and Ei- chenlaub 1997) to calculate a 100% minimum convex polygon (MCP) territory size estimate for each pair. In the same manner, we created MCP home range size es- timates of 1 1 Red-tailed Hawk territories mapped during the Snyder et al. study (H. Snyder unpubl. data) and of the 16 Red-tailed Hawk territories mapped by Santana and Temple (1984) to compare with our data. We report means and standard errors for our calcula- tions of territory sizes and overlap during each study pe- riod. We calculated the mean size for each territory that was used and also the proportion of the study area that comprised territories in all three periods. We used an analysis of variance to examine changes in territory sizes among the study intervals and an analysis of variance on ranks to examine changes in territorial overlap for those territories that were occupied consistently during all three studies. We used a paired Mest to examine differ- ences in territory size among territories detected in the 1980s study and 1998, but not those detected during the 1970s because the defined study area was smaller. We used a U.S. Geological Survey 30-m Digital Eleva- tion Model (DEM) to characterize topographic features of each home range. Using ArcView Spatial Analyst (En- vironmental Systems Research Institute, Inc.), we derived slope and aspect grids for the study area. These grids were used to calculate mean elevation, slope, and aspect for each home range. Mean aspect was the circular mean (Fisher 1993). We also created a 10-m contour map of the study area. Using the ArcView XTools extension (Oregon Department of Forestry), we calculated the lengths of each 10-m elevation contour within each home range. We obtained a measure of topographic complexity by dividing length of contours by home range size. Our measure of topographic complexity was analogous to the Land Surface Ruggedness Index (LSRI) proposed by Beasom et al. (1983). We used Pearson correlations and multiple regression to examine the relationship between elevation, slope, as- pect, topographic complexity, and territory size. We used a Kruskal-Wallis test to determine if years could be pooled for further analysis. We tested for differences m mean aspect between years using Watson’s test (Fisher 1993). All linear variables were examined for normality Variables that were not normally distributed were trans- formed using natural log transformations. Principal com- ponents analysis (PCA) was used as a variable reduction method if independent variables were highly correlated Several linear regression models were created. Mean as- pect, a circular variable, was treated as a linear model with cosine (aspect) and sine (aspect) as the indepen- dent variables (Fisher 1993). For all regressions, territory size was considered the response variable. We used Akai- ke’s Information Criterion (AIC) to select the best re- gression model (Burnham and Anderson 1998). Because we suspected that different features of the topography might have influenced territory size at different eleva- tions, territories were divided into high and low elevation groups using 700 m as an arbitrary breakpoint to divide 280 Boai. et al. VoL. 37, No. 4 the sample approximately in half. Regression models were reanalyzed in the manner described above using these two groups. All statistical analyses were conducted using SPSS for Windows 10.0.7 (SPSS, Inc.). Results Our results were consistent with those from the previous studies in terms of both number of terri- torial pairs and the individual territorial boundar- ies. We obtained adequate territory data for 16 pairs of Red-tailed Hawks in the study area. Mean size of defended Red-tailed Hawk territories during our study was 124.3 ± 12.0 ha (i.e., 1 pair/1.24 km^) . This was similar to our calculations of mean territory sizes of 139.0 ± 15.0 ha for 11 pairs from the 1970s study (Snyder unpubl. data) and 153.7 ± 15.8 ha for 16 pairs from the 1980s study (San- tana and Temple 1984) . Assuming territory size re- lated to breeding density, the density of breeding pairs of Red-tailed Hawks in the LEF was at the high end of the range of nesting densities reported for the species in 10 different studies (1 breeding pair/ 1.3-24. 9 km^; McGovern and McNurney 1986). In addition, there was substantial boundary overlap between defended territories. Territorial incursions by neighboring Red-tailed Hawks result- ed in territory overlap that was greatest in the 1970s study {x = 27.3 ± 8.8%), lowest in the 1980s study (x — 21.3 ± 3.3%), and in between the two during our 1998 study (x = 24.3 ± 4.2%), but these were not significant differences = 0.022, P = 0.989). In general, there was remarkably little temporal change in the spatial distribution, area, and bound- aries of Red-tailed Hawks among the study periods (Figs. 1-3) . Some territories, such as A and B, ap- peared to show slight increases and decreases among study periods (see Figs.). Other territories appeared to have disappeared altogether, but were replaced by new territories nearby. For example, territory I of the 1970s study (Fig. 1) may have shifted to become territory Q in the 1980s study (Fig. 2), but neither territory was occupied in 1998 (Fig. 3). Fikewise, territory K in the 1970s study (Fig. 1) and K in 1998 (Fig. 3) may have also heen R or P in the 1980s study (Fig. 2). Nine territories (A, B, C, D, E, F, G, H, and J; Figs. 1-3) that were detected in all three study pe- riods appeared to have been relatively consistent spatially. Although there were temporal fluctua- tions in the boundaries of these territories, over half of the area (x = 55.3 ± 2.9%) within each territory was common to that territory in all three studies. Furthermore, the territories did not tem- porally differ in sizes (^2 ^4 = 0.726, P = 0.494). There were some territories not detected or mapped during the 1970s study, due to a more re- stricted study area than in the 1980s study or our 1998 study. Three of these (L, M, and O) appeared spatially similar between the 1980s and 1998 (Figs. 2 and 3). Territory N in the 1980s study (Fig. 2) appeared to have dissolved into two territories (N and T) in 1998 (Fig. 3). When pooling T into N for 1998, there was no significant difference in size between the 1980s and 1998 territories at locations L, M, N, and O = 0.065, P = 0.952). None of the linear topographic variables dif- fered among years (Elevation, x ^2 ~ 2.384, P = 0.304; Slope, x% — F274, P — 0.529; Topographic Complexity, x ^2 ~ 1.636, P = 0.441; Aspect, x ^2 ~ 0.028, P = 0.986); therefore, we pooled all years (Table 1 ) . Territory size was transformed using the natural logarithm. All linear variables except ele- vation were negatively correlated with territory size (Elevation, r= 0.041, P= 0.793; Slope, r= —0.401, P = 0.008; Topographic Complexity, r = —0.406, P = 0.007). Slope and topographic complexity were strongly correlated with each other (r = 0.994, P < 0.001); therefore, they could not he used as in- dependent variables in the same regression model. A new variable was created based on component I of the PGA used to combine slope and topographic complexity. Because of the extremely low correla- tion of elevation and territory size, elevation was not used as an independent variable in any regres- sion model. Seven regression models were tested. Models 1- 3 were single-variable linear regressions of slope, topographic complexity, and the PGA variable, re- spectively. Model 4 was a linear regression with co- sine (aspect) and sine (aspect) as the independent variables. Models 5-7 were all possible pair-wise combinations of the four topographic variables. Size of territory was not independent of slope and aspect. Territory size decreased as slope in- creased. Generally, the smallest territories had as- pects facing southeast. The model with the best data fit predicts territory size using the formula: Territory size = g5. 6762-0. 038 Slope +0.1 395 Cosine (Aspect) -0. 1404 Sine (Aspect) The slope-aspect model received the smallest AIG value, and the best single-variable model was based on topographic complexity. The topographic vari- December 2003 Red-tailed Hawk Territories in Puerto Rico 281 Figure 1. Estimated spatial distribution of Red-tailed Hawk (RTHA) territories in the Luquillo Experimental Forest, Puerto Rico, during the early 1970s (H. Snyder unpubl. data). ables that influenced territory size varied by ele- vation. Low-elevation territories were most influ- enced by aspect and topographic complexity. High-elevation territories were most influenced by aspect. Discussion We suspect the temporal and spatial stability of Red-tailed Hawk territories in the LEF might be related to the hawks’ apparent use of permanent geographic features such as hilltops and ridgelines to delineate territory boundaries and corners. These geographic features might serve an impor- tant function by funneling the updrafts from the tradewinds. This allows Red-tailed Hawks on the east and northeast sides of the highest peaks and ridgelines to soar or hover in the strong updrafts, from which they can hunt and defend any part of the territory with a minimum expenditure of en- ergy. In addition, the relatively constant updrafts allow the Red-tailed Hawks to be on the wing from pre-dawn to twilight. This is especially true at high- 282 Boal et al. VoL. 37, No. 4 Figure 2. Estimated spatial distribution of Red-tailed Hawk (RTHA) territories in the Luquillo Experimental Forest, Puerto Rico, during the early 1980s, based on Santana and Temple (1984). er elevations where the updrafts should be stron- ger. Thus, the hawks are not constrained by time of day, as is the case in areas where soaring is de- pendent upon thermal updrafts. For example, Red-tailed Hawks occupying areas on the lee side of ridges usually gained and maintained altitude by extended flapping and circling rather than soar- ing. The role of topographic complexity in territory size is difficult to interpret. Increasing complexity may create more regions within a given territory in which updrafts occur, resulting in more areas in which a hawk can efficiently soar. Topographic complexity may also indicate a greater surface area within a territory than would be estimated by the MCP estimate, which is based on a flat surface area. However, we found a very high correlation between slope and topographic complexity. This might indicate that the LSRI method (Beasom et al. 1983) of measuring topographic complexity is December 2003 Red-tailed Hawk Territories in Puerto Rico 283 Figure 3. Estimated spatial distribution of Red-tailed Hawk (RTHA) territories in the Luquillo Experimental Forest, Puerto Rico, during January 1998 (this study). overly sensitive to slope, conveying little additional information about the area. There appeared to be little variability in territo- rial boundaries and areas among the three studies, despite a passage of 26 yr and forest damage as- sociated with Hurricane Hugo in 1989. The hur- ricane struck the LET study area with 4 hr of winds reaching as high as 227 km/hr (Scatena and Lar- sen 1991). Resulting damage to the forest was deemed so severe that forest composition and dy- namics were affected for several years (Brokaw and Grear 1991). Removal of canopy trees by the hur- ricane resulted in numerous openings in the forest canopy (Brokaw and Grear 1991, Scatena and Lar- sen 1991, Waide 1991). Openings in the canopy might have facilitated hunting by Red-tailed Hawks and, although there are no supportive data, might have resulted in an increase in Red-tailed Hawks in the LEE. If any increase occurred, however, it was likely short-lived, as the Red-tailed Hawk den- 284 Boal et al. VoL. 37, No. 4 Table 1. Mean elevation (m), slope and aspect (de- grees), and topographic complexity (m/ha) of Red-tailed Hawk territories, Luquillo Experimental Forest, Puerto Rico. Data derived from 1972-75 (Snyder et al. 1987), 1981-83 (Santana and Temple 1984), and 1998 (this study) were pooled. Variable Mean Stan- dard Error Mini- mum Maxi- mum Elevation 671.1 23.2 300.9 956.0 Slope 19.5 0.6 12.8 28.2 Aspect 107.1 14.4 12.0 289.0 Topographic complexity 395.5 12.3 269.0 583.8 sity in 1998 was not different than the pre-hurri- cane density (Santana and Temple 1984, Snyder et al 1987). The 1970s study (Snyder et al. 1987) and 1980s study (Santana and Temple 1984) did not report the number of days of Red-tailed Hawk observa- tion. However, it is obvious our study of 15 days was quite limited in duration compared to these earlier investigations. The differences we found in some territory boundaries might be attributable to the time constraints of our study compared to the earlier studies (Santana and Temple 1984, Snyder et al. 1987), and that our survey was conducted in January, which is at the start of the nesting cycle for resident Red-tailed Hawks and when they are presumably most aggressive in territorial defense (Snyder et al. 1987). However, there was no statis- tical difference in the territory sizes or boundary overlap among the study periods. Thus, we believe our data accurately depict territory sizes, bound- aries, and Red-tailed Hawk density at the onset of the 1998 breeding season. There appeared to be a number of individual Red-tailed Hawks that were unattached to a mate or territory passing through the study area and they traversed the defended territories by main- taining high altitudes. Territorial residents would occasionally soar upward, but remained below the intruder as it passed over the area. As the intruder approached a boundary, the adjacent resident Red- tailed Hawks would rise up toward the intruder. In this way, a high-altitude intruder would be “escort- ed” across the study area by each successive pair of resident hawks. These encounters appeared to be non-aggressive as long as the intruder main- tained a high altitude and continued moving across the area. Only when an intruder began de- scending or circling were chases initiated. None of the high-altitude intruders were observed diving into known territories. Rather, they invariably “van- ished” into the distance after having passed over several occupied territories. Therefore, we as- sumed that these intruders were not, in fact, adja- cent or distant neighbors who were cruising above the landscape. In addition to an unknown number of adult Red-tailed Hawks that flew over the terri- tories, we observed at least two immature Red- tailed Hawks that appeared to be living inconspic- uously among the territories in the study area. Immature Red-tailed Hawks were rarely observed soaring. Rather, they were observed moving low above the canopy of the forest, and they quickly dove out of sight into the canopy when detected and pursued by a resident. Territorial behavior among birds can lead to generally even distribution of individuals within a population (Newton 1998). High site-fidelity among individuals in a population might lead to relatively long-term spatial stability of territory ar- eas. Furthermore, stability of a population over sev- eral years suggests that densities might be limited by competition for territorial space (Newton 1991). Our data, combined with that of Santana and Temple (1984) and Snyder et al. (1987), sug- gest there was a high degree of temporal stability in the density and spatial distribution of Red-tailed Hawks in the LEF that may be linked to geographic features. Furthermore, the temporal stability of the spatial distribution of Red-tailed Hawk territories in the study area leads us to believe the area is at or near saturation. Acknowledgments We thank J. Wunderle, the International Institute of Tropical Forestry, and the USDA Forest Service, Carib- bean National Forest for facilitating this research. R. Smith, H. Garner, N. McIntyre, J. Wunderle, and an anonymous reviewer provided constructive comments on an earlier draft of the manuscript. Funding was provided by the International Institute of Tropical Forestry. Literature Cited BEASOM, S.L., E.P. WiGGERS, AND J.R. GlARDINO. 1983. A technique for assessing land surface ruggedness. J. Wildl. Manage. 47:1163-1166. Brokaw, N.V.L. AND J.S. Grear. 1991. Forest structure before and after Hurricane Hugo at three elevations in the Luquillo Mountains, Puerto Rico. Biotropica 25: 386-392. Brown, S., A.E. Lugo, S. Silander, and L. Liegel. 1983. December 2003 Red-tailed Hawk Territories in Puerto Rico 285 Research history and opportunities in the Luquillo Experimental Forest. U.S. For. Serv. Gen. Tech. Rep. SO-44. Southern Forest Experiment Station, New Or- leans, LA U.S.A. Burnham, K.R and D.R. Anderson. 1998. Model selec- tion and inference: a practical information-theoretic approach. Springer-Verlag, New York, NY U.S.A. Ewel, JJ- AND J.L. Whitmore. 1973. The ecological life zones of Puerto Rico and the U.S. Virgin Islands. U.S. Forest Service, Institute of Tropical Forestry, Res. Pa- per ITF-18. Rio Piedras, Puerto Rico. Fisher, N.I. 1993. Statistical analysis of circular data. Cambridge Univ. Press, Cambridge, U.K. GarcIa-Martino, A.R., G.S. Warner, F.N. Scatena, and D.L. Crvco. 1996. Rainfall, runoff, and elevation re- lationships in the Luquillo Mountains of Puerto Rico. Canbb.J. Sci. 32:413-424. Hooge, PN. and B. Eichenlaub. 1997. Animal move- ment extension to Arc View Ver. 1.1. Alaska Biological Science Center, U.S. Geological Survey, Anchorage, AK U.S.A. Howell, J., B. Smith, J.B. Holt, Jr., and D.R. Osborne. 1978. Habitat structure and productivity in Red-tailed Hawks. Bird Banding 49:162-1^1. Janes, S.W. 1984. Influences of territory composition and interspecific competition on Red-tailed Hawk repro- ductive success. Ecology 65:862-870. McGovern, M. and J.M. McNurney. 1986. Red-tailed Hawk nest densities, Colorado. Raptor Res. 20:43-45. Newton, I. 1991. Population limitation in birds of prey: a comparative approach. Pages 3-21 in C.M. Perrins, J.-D. Lebreton, and G.J.M. Hirons [Eds.], Bird popu- lation studies: relevance to conservation and manage- ment. Oxford Univ. Press Inc., New York, NYU.S.A. . 1998. Population limitation in birds. Academic Press, San Diego, CA U.S.A. Powell, R.A. 2000. Animal home ranges and territories and home range estimators. Pages 67-110 in L. Boi- tani and T.K. Fuller [Eds.], Research techniques m animal ecology. Columbia Univ. Press, New York, NY U.S.A. Preston, C.R. and R.D. Beane. 1993. Red-tailed Hawk {Buteo jamaicensis) . In A. Poole and F. Gill [Eds.], The Birds of North America, No. 52. The Birds of North America, Inc., Philadelphia, PA U.S.A. Santana, E. and S.A. Temple. 1984. Management rec- ommendations for the endangered Puerto Rican Par- rot and the Red-tailed Hawk in the Luquillo Experi- mental Forest. Final Report Cooperative Agreement No. 29:80-393. AND . 1988. Breeding biology and diet of Red-tailed Hawks in Puerto Rico. Biotropica 20:151- 160. Scatena, F.N. and M.C. Larsen. 1991. Physical aspects of Hurricane Hugo in Puerto Rico. Biotropica 23:317- 323. Snyder, N.F.R., J.W. Wiley, and C.B. Kepler. 1987. The parrots of Luquillo: natural history and conservation for the Puerto Rican Parrot. Western Foundation of Vertebrate Zoology, Los Angeles, CA U.S.A. AND H.A. Snyder. 1991. Birds of prey: natural his- tory and conservation of North American raptors Voyageur Press, Stillwater, MN U.S.A. Waide, R.B. 1991. The effect of Hurricane Hugo on bird populations in the Luquillo Experimental Forest, Puerto Rico. Biotropica 23:475-480. Received 8 November 2002; accepted 23 July 2003 J Raptor Res. 37(4):286-291 © 2003 The Raptor Research Foundation, Inc. WINTERING RAPTOR USE OE HYBRID POPLAR PLANTATIONS IN NORTHEASTERN OREGON Brian W. Moser^ Potlatch Corporation, Hybrid Poplar Program, PO. Box 38, Boardman, OR 97818 U.S.A. G. Keith Hilpp Potlatch Corporation, PO. Box 1323, Lewiston, ID 83501 U.S.A. Abstract. — We studied wintering raptor use of hybrid poplar {Populus spp.) plantations in comparison to surrounding cover types in the Columbia Basin of northeastern Oregon. Diurnal raptors were sur- veyed in shrub-steppe, poplar plantations, and irrigated croplands. Logistic regression analyses suggested that the three most common raptors. Red-tailed Hawks {Buteo jamaicensis), American Kestrels {Falco sparverius) , and Northern Harriers {Circus cyaneus) were associated with croplands, interiors of 1-yr-old plantations, and plantation edges. Shrub-steppe was also selected as a significant predictor of sites with American Kestrels. The best model for the Northern Harrier also included the interior of 2-yr-old plantations, but excluded croplands and edges of older plantations. Plantations and plantation edges appeared to be used by wintering raptors disproportionately to their availability. Our data suggest that maximizing plantation edges and managing for a variety of plantation ages within this landscape will likely provide suitable habitat for wintering raptors in this region. Kty Words: American Kestrel, Falco sparverius; Northern Harrier, Circus cyaneus; Red-tailed Hawk, Buteo jamaicensis; cropland', hybrid poplar, Populus spp.', shrub-steppe. USO DE PIANTACIONES DE ALAMO HIBRIDO POR RAPACES DURANTE EL INVIERNO EN EL NORORIENTE DE OREGON Resumen. — Estudiamos el uso por rapaces durante el invierno de plantaciones de alamo hibrido {Po- pulus spp.) en comparacion a los tipos de cobertura de los alrededores en la cuenca del Columbia en el nororiente de Oregon. Las rapaces diurnas fueron estudiadas en plantaciones, campos irrigados y habitats de estepas arbnstivas. El analisis de regresion logistica sugiere que la mayoria de rapaces estan asociadas con los campos, los interiores de las plantaciones de un afio, y los hordes de las plantaciones. La estepa arbustiva fue seleccionada ademas como un pronosticador efectivo del cernicalo {Falco spar- verius). El mejor modelo para el aguilucho norteho {Circus cyaneus) incluyo ademas los interiores de plantaciones de 2 ahos de edad, pero excluyo los campos y hordes de plantaciones mas viejas. Las plantaciones parecieron ser un habitat importante para rapaces invernantes, especialmente como sitios de percha. Nuestros datos sugieren que para la mayoria de especies, maximizar los hordes de las plan- taciones y proveer una variedad de edades en la plantacion dentro de este paisaje probablemente proporcionara el mejor habitat para las rapaces durante el invierno. [Traduccion de Cesar Marquez] The development of short-rotation woody crop (SRWC) plantations in North America is becoming increasingly common with the forest-products in- dustry due to the fast-growing nature of these trees. Nearly 80 000 ha of SRWC plantations have been established in the United States, and over 69 million ha have been identified as suitable for ^ Present address: Potlatch Corporation, P.O. Box 1388, Lewiston, ID 83501; e-mail address: brian.moser® potlatchcorp.com SRWC plantations (Alig et al. 2000). Although a variety of tree species are grown in these planta- tions, hybrid poplars {Populus spp.) are the most commonly planted species in the Pacific North- west. In the past 15 yr, over 28 000 ha of poplar plantations have been established in the Pacific Northwest to produce fiber for paper and dimen- sional lumber products (Heilman et al. 1995). Pop- lars are also used in conservation plantings, as buff- ers along riparian areas, and in phytoremediation projects across North America (Isebrands 2000) . 286 December 2003 Raptor Use of Poplar Plantations 287 Little is known about the habitat value of these plantings for raptors. However, Bogliani (1988), Prater (1993), Bogliani et al. (1994), Sergio and Bogliani (2000) have examined bird use of poplar plantations in Europe. Bogliani et al. (1994) and Sergio and Bogliani (2000) reported that poplar plantations supported nesting Eurasian Hobby {Falco suhbuteo) densities of up to one pair per 50 km^. Prater (1993) also commented that poplar plantations provided hunting habitat for the Barn Owl {Tyto alba) in Great Britain. Moreover, Schmitz (1986) demonstrated that Common Buzzards {Bu- teo buteo) used poplar plantations in Belgium, al- though at lower relative densities than other bird species. Fewer studies have examined raptor use of pop- lar plantations in North America. Moser (2002) found that nine Northern Saw-whet Owl {Aegolius acadicus) pairs nested successfully in poplar plan- tations when 25 nest boxes were provided. Chris- tian et al. (1997) studied bird use of poplar plan- tations in the Midwest in comparison to bird use of surrounding croplands and natural forests. They reported Red-tailed Hawks {Buteo jamaicensis) used forests adjacent to plantations, but were not found in the plantations. Because poplar plantations commonly replace cropland, and sometimes native habitat, it is im- portant that we understand the use of these plan- tations by wildlife. The objective of our study was to determine which cover types were the best pre- dictors of raptor abundance in a landscape domi- nated by shrub-steppe, poplar plantations, and croplands in northeastern Oregon. Study Area The 20 105 ha study area was a complex of poplar plan- tations and irrigated croplands, surrounded by shrub- steppe vegetation (Franklin and Dyrness 1988) in the Co- lumbia Basin of northeastern Oregon. Shrub-steppe, plantation, and cropland cover types comprised 47%, 35%, and 18% of the study area, respectively. The area contained relatively high densities of primitive roads in plantations and croplands (>1.9 km/km^), and lower densities in shrub-steppe (<1 km/km^). Topography var- ied from flat to slightly undulating and elevations ranged from 150-250 m, Mean annual precipitation was ca, 22 cm (Ruffner 1978). Shrub-steppe vegetation varied depending on soil type (Franklin and Dyrness 1988) . The Quincy sand soil types were dominated by indian ricegrass {Oryzopsis hymenoi- des), needle-and-thread (Stipa comata), and antelope bit- terbrush {Purshia tridentata). The finer soils were domi- nated by basin big sagebrush {Artemisia tridentata), bluebunch wheatgrass {Pseudoregnaria spicatum), Idaho fescue {Festuca idahoensis), and Sandberg’s bluegrass {Poa sandhergii). This vegetation was also found in undis- turbed, unfarmed corners (<2 ha each) surrounding ag- ricultural circles. Corners that had been mowed or cul- tivated were usually dominated by invader species such as cheatgrass {Bromus tectorum) and Russian thistle {Sal- sola kali), or planted with a cover crop such as wheat Perch sites were relatively scarce in this cover type, and when available consisted of power poles, fence posts, and shrubs. Drip-irrigated plantations were first established in 1994, primarily on former croplands irrigated by center- pivot systems. Plantations representing six age classes of trees (1-6-yr-old) were available for study. All trees were planted on ca. 2.5 X 3 m spacing and scheduled for har- vest within 8—12 yr. Plantations differed in both tree height and understory composition depending on age. On average, poplars grew 3-5 m/yr. During the first 2-3 yr of growth, understory vegetation was often dense (>90% cover). Common understory plants included the native horseweed {Conyza canadensis) and redroot pig- weed {Amaranthus palmeri) , and several introduced invad- ers, including Russian thistle, tumble mustard {Sisym- brium altissimum), and cheatgrass. Each plantation was sprayed with an herbicide-mixture of glyphosate and 2,4- D, and cultivated once during yr one and two. However, most competing plants quickly recolonized given the fa- vorable growing conditions, and impacts to rodent pop- ulations were short-term (Moser et al. 2002). Complete canopy closure occurred following yr three. At this time, the understory became nearly devoid of live plants (<5% cover) and was dominated by organic material primarily composed of leaf litter. Perch sites were abundant in all plantation age classes. Agricultural fields in the study area contained a cover crop such as winter wheat, were fallow with wheat or corn stubble, or were planted with a perennial crop such as alfalfa or grass. These fields were not irrigated during the study period. Perch sites were more abundant than m shrub-steppe, but less abundant than plantation cover types and included hay stacks, power poles, small trees, and irrigation equipment. Methods Cover types were surveyed with equal intensity, and in- cluded shrub-steppe, tree plantations by age-class (1-6- yr-old), and croplands. Tree plantations were further stratified into interior and edge categories, with the ex- ception of 6-yr-old plantations, which due to limited area had only edge transects. Plantation-interior categories were bordered on all sides by plantations. Plantation- edge categories were defined as the convergence of plan- tations and either shrub-steppe or cropland cover types These habitat cover types used in the analysis are de- scribed in Table 1. We conducted road surveys through the center of each of the cover type categories once per week during Janu- ary and February, for a total of eight surveys in 1999 and nine in 2000. Primitive road sections were classified as either containing powerlines or not, and road sections were randomly selected for surveys until each cover type had equal amounts of powerline-influenced road sec- tions. Approximately 6.4 km of transect was selected as 288 Moser and Hilpp VoL. 37, No. 4 Table 1. Indicator variables used in the development of logistic regression models for wintering raptors in north- eastern Oregon, 1999 and 2000. Variable Description YR 1999 or 2000 NAT Shrub-steppe habitat AG Irrigated cropland 11 Interior of 1-yr-old plantations 12 Interior of 2-yr-old plantations 13 Interior of 3-yr-old plantations 14 Interior of 4-yr-old plantations 15 Interior of 5-yr-old plantations El Edge of 1-yr-old plantations E2 Edge of 2-yr-old plantations E3 Edge of 3-yr-old plantations E4 Edge of 4-yr-old plantations E5 Edge of 5-yr-old plantations E6 Edge of 6-yr-old plantations four 1.6 km sections of road in each of the cover types. Transects were located ^0.8 km apart to reduce the prob- ability of sampling the same bird twice (Fuller and Mosh- er 1987). Birds detected within 100 m of the transect were assumed to be using that cover type. Surveys were conducted between sunrise and 1100 H. Successive sur- veys were conducted in opposite directions to reduce any time-of-day bias. Surveys were conducted from a vehicle by two trained observers (all observers had a minimum of a B.S. in Wildlife Biology and were trained formally during a 5-d period to identify local raptors and to con- duct surveys) to identify local raptors while using binoc- ulars and spotting scopes to aid in raptor identification. The vehicle was traveling continuously at 10-30 km/hr, depending on visibility within each cover type. Raptor detection rates were influenced by both survey effort (i e., time) and visibility. In general, visibility was greatest m shrub-steppe and croplands, and decreased with in- creasing plantation age. Therefore, we adjusted vehicle speed so that it was greater in open cover types, and slow- er in dense cover types. Although this methodology re- sulted in greater survey time in older plantations, we felt any survey effort bias was appropriately offset by the low- er detection rates of birds in the heavier cover types (Millsap and LeFranc 1988). We stopped the vehicle to identify birds when necessary (Fuller and Mosher 1987). Surveys were not conducted during inclement weather (Ralph et al. 1993). All birds, whether perched or flying, were recorded and included in the analyses as using a particular cover type for two reasons. First, although most of the species we studied hunt primarily from perches, many of them will also hunt while flying. Two of the three most com- mon raptors we observed (Red-tailed Hawk and Ameri- can Kestrel) hunt from both a perch and while flying (Preston and Beane 1993, Smallwood and Bird 2002). Furthermore, the Northern Harrier {Circus cyaneus) hunts almost exclusively while flying (MacWhirter and Bildstein 1996). Second, we were not always able to de- termine whether we flushed a bird as we approached its perch, or whether it was flying before we detected it. Therefore, we assumed that if a bird was detected perched in or flying through a cover type, that it was using that cover type. Chi-square goodness-of-fit analysis of the count distri- bution indicated the data were not normally distributed, but rather conformed to the geometric distribution {P > 0.05). Attempts to transform the data to a normal distri- bution failed due to the low-frequency of the transect counts, the discrete nature of count data, and the prev- alence of transects with no raptor detections. Thus, the count data for each species were converted to a pres- ence/absence variable, and multiple logistic regression was used to determine the elements of the experimental design that contributed significantly to raptor presence. Logistic regression with indicator variables provides in- sight into the magnitude of effects of individual habitat attributes on raptor presence through the relative value of the coefficient estimate. Coefficient estimates with higher values indicate habitats with a higher probability of raptor presence. Sufficient observations existed to de- velop individual regression models for Red-tailed Hawk, Northern Harrier, and American Kestrel. Count data were coded according to habitat attributes using indica- tor variables (Table 1). Additionally, a temporal variable (YR) was included to determine the significance of year to the survey as this was an initial element of the exper- imental design (Table 1). Because raptor counts were generally higher in 1999 than in 2000, the YR variable was coded as 1 for 1999, and 0 for 2000. Thus, each anal- ysis of habitat associations contained 14 indicator vari- ables. Parameter estimates were developed using Maxi- mum Likelihood estimation (StatSoft, Inc. 1995). Indicator variables for P > 0.05 were removed individu- ally until only significant variables remained in the mod- el. All analyses were performed with Statistica for Win- dows (StatSoft, Inc. 1995). The classification ability of each model was tested by randomly sub-sampling 25% of the original data. The predicted values for each model were then compared to the observed values for this sub-sample. It should be not- ed that these data were used in the original model de- velopment, and cannot be considered independent. Nev- ertheless, they still provided some information on the classification ability of the models. Results We detected 374 diurnal raptors of 12 species during 1999 and 2000 (Table 2). The most com- mon raptors were Red-tailed Hawk, Northern Har- rier, and American Kestrel. Species detected at rel- atively low rates included Rough-legged Hawk {Buteo lagopus). Ferruginous Hawk {B. regalis), Northern Goshawk (Accipiter gentilis), Cooper’s Hawk {A. cooperii), Sharp-shinned Hawk (A. stria- tus). Golden Eagle {Aquila chrysaetos). Prairie Fal- con {Falco mexicanus), Merlin (K columbarius) , and Snowy Owl (Nyctea scandiaca). All but the Prairie Falcon were observed at least once within a plan- December 2003 Raptor Use of Poplar Plantations 289 Table 2. Total raptor detections by cover type in northeastern Oregon, 1999 and 2000. Cover Type^ Percent Total Perched NAT AG El E2 E3 E4 E5 E6 11 12 13 14 15 Total 11 49 17 21 35 57 68 42 49 11 6 6 2 374 89.2 Red-tailed Hawk 1 21 5 4 15 28 44 20 6 2 0 2 1 149 86.8 Northern Harrier 1 6 7 6 1 10 6 2 27 7 4 3 0 80 26.6 American Kestrel 3 17 2 11 11 9 9 6 7 0 0 0 0 75 89.5 Rough-legged Hawk 4 2 1 0 5 3 5 2 1 0 0 0 0 23 87.3 Cooper’s Hawk 0 0 0 0 0 3 0 0 2 1 1 0 0 7 100.0 Ferruginous Hawk 0 0 0 0 0 0 0 6 0 0 0 0 0 6 83.3 Golden Eagle 1 0 1 0 0 0 1 0 2 0 0 0 0 5 100.0 Sharp-shinned Hawk 0 0 0 0 0 0 1 1 1 0 0 0 0 3 100.0 Snowy Owl 0 1 0 0 0 0 0 0 1 0 0 0 0 2 100.0 Merlin 1 0 0 0 1 0 0 0 0 0 0 0 0 2 100.0 Prairie Falcon 0 1 0 0 0 0 0 0 0 0 0 0 0 1 100.0 Northern Goshawk 0 0 1 0 0 0 0 0 0 0 0 0 0 1 100.0 Unidentified raptor 0 1 0 0 2 4 2 5 2 1 1 1 1 20 85.5 ^ See Table 1 for description of cover types. Table 3. Logistic regression models used to identify cov- er types significant to raptor presence in northeastern Oregon, 1999-2000. All regression models are significant (P< 0.001). Dependent Variable Inter- cept Indi- cator Vari- able Coeffi- cient P Red-tailed Hawk -3.92 AG 2.89 <0.001 11 1.58 0.001 El 1.54 0.003 E2 1.45 0.008 E3 2.37 <0.001 E4 3.24 <0.001 E5 3.74 <0.001 E6 3.46 <0.001 Northern Harrier -3.25 11 2.22 <0.001 12 0.93 0.008 El 1.07 0.002 E2 1.23 <0.001 E4 1.23 <0.001 American Kestrel -5.40 NAT 1.89 0.004 AG 3.55 <0.001 11 2.81 <0.001 E2 3.32 <0.001 E3 3.23 <0.001 E4 2.96 <0.001 E5 3.10 <0.001 E6 3.79 <0.001 YR 0.78 <0.001 ’ See Table 1 for description of indicator variables. tation. Most of the raptors detected (except for Northern Harriers) were observed on a perch (Ta- ble 2). Logistic regression models developed for Red- tailed Hawk, Northern Harrier, and American Kes- trel, were all significant (P < 0.001; Table 3). The cover type categories selected as contributors to predictability of Red-tailed Hawk presence were AG, II, and E1-E6 (P < 0.008; Table 3). Cover type categories selected for the Northern Harrier mod- el included AG, 11—12, and El— E2, and E4 (P = 0.008; Table 3). The cover type categories selected as contributors to predictability of American Kes- trel presence were NAT, AG, 11, and E1-E6 (P = 0.004; Table 3). The 13-16 cover type categories were not selected for any model (P > 0.05) . Survey year was a significant component for the American Kestrel model only. More Kestrels were detected in 1999 than in 2000 (P < 0.001; Table 3). The in- dicator variables selected for inclusion or exclusion in the logistic models correctly predicted raptor presence 92% of the time within 10% of the ob- served rate. Variables that did not classify as well included AG and E4 for the Red-tailed Hawk mod- el, and 11 for the Northern Harrier model (Table 4). Discussion Although we detected 12 species of raptors in this study, only three were detected frequently enough to be considered common. All of the hab- 290 Moser and Hilpp VoL. 37, No. 4 Table 4. Predicted versus observed classification rates of logistic regression models for Red-tailed Hawk (RTHA), Northern Harrier (NOHA), and American Kestrel (AMKE) presence in various cover types. Indica- TOR Vari- able Predicted Rate Observed Rate RTHA NOHA AMKE RTHA NOHA AMKE NAT 0.00 0.00 0.03 0.00 0.00 0.00 AG 0.26 0.00 0.14 0.07 0.00 0.07 11 0.09 0.26 0.07 0.13 0.38 0.13 12 0.00 0.09 0.00 0.00 0.05 0.00 13 0.00 0.00 0.00 0.00 0.07 0.00 14 0.00 0.00 0.00 0.06 0.00 0.00 15 0.00 0.00 0.00 0.00 0.00 0.00 El 0.09 0.10 0.00 0.00 0.11 0.00 E2 0.08 0.12 0.11 0.07 0.14 0.07 E3 0.18 0.00 0.10 0.18 0.00 0.09 E4 0.34 0.12 0.08 0.50 0.13 0.06 E5 0.46 0.00 0.09 0.45 0.00 0.09 E6 0.39 0.00 0.17 0.42 0.00 0.25 ' See Table 1 for description of indicator variables. itats studied were used by at least one raptor spe- cies. However, raptor distribution was most affect- ed by croplands and poplar plantations in this study. In fact, only American Kestrel distribution was significantly affected by native shrub-steppe habitat. Our results are not surprising, as a number of other studies have reported relatively high num- bers of certain raptor species in agricultural land- scapes (Parker and Campbell 1984, Craig et al. 1986, Meunier et al. 2000), probably due to both a large number of available perch sites (e.g., utility poles) as well as relatively high densities of small mammal prey in these settings. The distribution of utility lines and poles within treatment categories in our study was intentionally made equal so as not to confound the results. Nevertheless, plantations still provided significantly more perch sites than shrub-steppe or even cropland cover types. Thus, it is not surprising that most raptors were more commonly detected in or near plantations. Fur- thermore, prey densities are relatively high in both croplands and young poplar plantations (Christian et al. 1997, Moser et al. 2002). Because most Red- tailed Hawks and American Kestrels were detected on perches, it is likely they were hunting in those habitats (Preston and Beane 1993, Smallwood and Bird 2002) . Likewise, most Northern Harriers were detected flying, again suggestive of hunting behav- ior for this species (MacWhirter and Bildstein 1996) . Hanowski et al. (1997) found that surrounding cover types affected songbird use of hybrid poplar plantations in the midwestern U.S. We believe this phenomenon to be true in our study as well. Rap- tor detections were generally greatest along the edges of plantations when compared to the inte- riors, with the exception of the Northern Harrier which commonly hunts in open, brushy habitats, similar to those provided by tbe younger planta- tions. Most of the birds studied were not well adapted to hunting in the dense plantation habi- tats, with the exception of the Accipiter spp. The edge habitat created by the convergence of plan- tations with cropland and shrub-steppe habitats provided ideal hunting areas by providing perch sites near open habitat suited for prey species such as rodents. Raptors contribute to important ecosystem pro- cesses, such as predator-prey dynamics (Hanski et al. 1991), and thus may be important in dampen- ing the effects of nuisance rodent outbreaks (Kor- pimaki and Norrdahl 1991). In addition, these birds may be indicators of ecosystem quality due to their high-trophic-level status. Due to the ecologi- cal and economic importance of this bird assem- blage, it is important that plantation managers con- sider these birds when designing and managing plantations. Due to the affinity of raptors toward both the interiors and edges of plantations, main- taining a diversity of plantation ages and creating within-plantation heterogeneity (Hanowski et al. 1997) will provide more structural diversity and edge, and thus likely create more favorable habitat for wintering raptors. Acknowtedgments Financial support for this research was provided by Pot- latch Corporation and Boise-Cascade. Study sites were also provided by the U.S. Naval Weapons Systems Train- ing Facility. We thank C.A. Duberstein, M.C. Nowak, and A. Rice for helping with the raptor surveys. Literature CriED Alig, R.J., D.M. Adams, B.A. McCaru, and P.J. Ince. 2000. Economic potential of short-rotation woody crops on agricultural land for pulp fiber production in the United States. For. Prod. J. 50:67-74. Bogliani, G. 1988. Density and habitat choice of nesting birds in poplar plantations. Riv. Ital. Ornitol. 58:129- 141. , F. Barbieri, and E. Tiso. 1994. Nest-site selection December 2003 Raptor Use of Poplar Plantations 291 by the hobby {Falco subbuteo) in poplar plantations in northern Italy./. Raptor Res. 28:13-18. Christian, D.P., P.T. Colons, J.M. Hanowski, and G.J. Nieml 1997. Bird and small mammal use of short- rotation hybrid poplar plantations. / Wildl. Manage. 61:171-182. Craig, E.H., T.H. Craig, and L.R. Powers. 1986. Habitat use by wintering Golden Eagles and Rough-legged Hawks in southeastern Idaho. Raptor Res. 20:69-71. Eranklin, J.F. AND C.T. Dyrness. 1988. Natural vegetation of Oregon and Washington. Oregon State Univ. Press, Corvallis, OR U.S.A. Fuli-ER, M.R. and J.A. Mosher. 1987. Raptor survey tech- niques. Pages 37-65 in B.A. Giron-Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor Management Techniques Manual. National Wildlife Federation Scientific and Technical Series No. 10, Washington, DC U.S.A. Hanowski, J.M., G.J. Niemi, and D.P. Christian. 1997. Influence of within-plantation heterogeneity and sur- rounding landscape composition on avian communi- ties in hybrid poplar plantations. Conserv. Biol. 11:936- 944. Hanski, L, L. Hansson, and H. Henttonen. 1991. Spe- cialist predators, generalist predators, and the micro- tine rodent cycle. J. Anim. Ecol. 60:353-367. Heilman, P.E., R.F. Stettler, D.P. Hanley, and R.W. Carkner. 1995. High yield hybrid poplar plantations in the Pacific Northwest, PNW Extension Bulletin No. 356. Washington State Univ. Coop. Ext., Pullman, WA U.S.A. Isebrands, J.G. 2000. Activities related to poplar and wil- low cultivation and utilization. National Poplar Com- mission of the USA report to the International Poplar Commission for the period 1996-2000. Food and Ag- riculture Organization of the United Nations, Rome, Italy. Korpimaki, E. and K. Norrdahl. 1991. Do breeding no- madic avian predators dampen population fluctua- tions of small mammals? Oikos 62:195-208. MacWhirter, R.B. and K.L. Bild.stein. 1996. Northern Harrier {Circus cyaneus) . In A. Poole and F Gill [Eds.], The birds of North America, No. 210. Academy of Natural Sciences, Philadelphia, PA, and American Or- nithologists’ Union, Washington, DC U.S.A. Meunier, F.D., C. Verheyden, and P. Jouventin. 2000 Use of roadsides by diurnal raptors in agricultural landscapes. Biol. Conserv. 92:291-298. Millsap, B.A. and M.N. LeFranc, Jr. 1988. Road transect counts for raptors: how reliable are they? J. Raptor Res. 22:8-16. Moser, B.W. 2002. Reproductive success of Northern Saw-whet Owls nesting in hybrid poplar plantations Northwest Sci. 76:353-355. , MJ. PiPAS, G.W. Witmer, and R.M. Engeman 2002. Small mammal use of hybrid poplar plantations relative to stand age. Northwest Sci. 76:158-165. Parker, R.E. and E.G. Campbell. 1984. Habitat use by wintering birds of prey in southeastern Arizona. West Birds 15:175-183. Prater, A.J. 1993. Poplars and wildlife. Quart. J. For. 87. 314-318. Preston, C.R. and R.D. Beane. 1993. Red-tailed Hawk {Buteo jamaicensis) . In A. Poole and F Gill [Eds.], The birds of North America, No. 52. Academy of Natural Sciences, Philadelphia, PA, and American Ornitholo- gists’ Union, Washington, DC U.S.A. Ralph, C.J., G.R. Geupel, P. Pyi.e, T.E. Martin, and D.F DeSante. 1993. Handbook of field methods for mon- itoring landbirds. U.S. For. Serv. Gen. Tech. Rep PSW-144. Albany, CA U.S.A. Ruffner, J.A. 1978. Climates of the United States. Vol. 2 Gale Research Company, Detroit, MI U.S.A. Schmitz, L. 1986. Breeding and wintering avifauna in West Hesbaye poplar plantations. Aves 23:81-120. Sergio, F. and G. Bogliani. 2000. Hobby nest-site selec- tion and productivity in relation to intensive agricul- ture and forestry./. Wildl. Manage. 64:637-646. Smallwood, J.A. and D.M. Bird. 2002. American Kestrel {Falco sparverius). In A. Poole and F. Gill [Eds.], The birds of North America, No. 602- Academy of Natural Sciences, Philadelphia, PA, and American Ornitholo- gists’ Union, Washington, DC U.S.A. StatSoft, Inc. 1995. Statistica for Windows. Vol. 3: Sta- tistics II. 2nd Ed. StatSoft, Inc., Tulsa, OK U.S.A. Received 20 February 2003; accepted 28 August 2003 Associate Editor; Ian G. Warkentin J Raptor Res. 37(4):292-298 © 2003 The Raptor Research Foundation, Inc. MORTALITY OF WINTERING OSPREYS AND OTHER BIRDS AT AQUACULTURE FACILITIES IN COLOMBIA MarcJ. Bechard^ Department of Biology, Boise State University, Boise, ID 83725 US. A. Cesar Marquez-Reyes^ Alexander von Humboldt Institute, Carrera 7, No. 35-20, Bogota, Colombia Abstract. — Ospreys {Pandion haliaetus) winter throughout the country of Colombia. Recoveries of band- ed Ospreys indicate that many are shot in the country with the number increasing since the 1970s. The increased incidence of shooting has coincided with the development of aquaculture facilities in Colom- bia that raise tilapia {Oreochromis spp.). Because these facilities typically lose production to birds such as Ospreys that depredate fish, we conducted a survey of 83 facilities in three states or departments in Colombia in 2001 to determine the species of birds that take fish at aquaculture facilities and the numbers that are killed each year. Our results showed that bird depredation occurs at aquaculture facilities throughout the country, but mostly in the southern portion of the country in the department of Huila. Facility managers reported shooting Ospreys in all three departments with as few as five individuals shot annually in Antioquia in northern Colombia to as many as 270 shot annually in Huila. In addition, facility managers reported shooting nine other species of birds including Green Kingfishers {Chloroceryle americana), Great Kiskadees {Pitangus sulphuratus) , Snowy Egrets {Egretta thula). Great Egrets {Casmerodius albus), Olivaceous Cormorants (Phalacrocorax olivaceus), Black-crowned Night-Herons {Nyc- ticorax nycticorax) , White-necked Herons {Ardea cocoi). Cattle Egrets {Bubulcus ibis), and Striated Herons {Butorides striatus), for a total estimate of >9000 birds shot in the three departments annually. A number of alternative methods to shooting had been used to reduce losses to birds including the installation of netting, overhead wires, scarecrows, and noise making devices, but, neither these methods nor shooting, were effective in deterring avian predators. Keywords: Osprey, Pandion haliaetus; aquaculture facilities] mortality] Oreochromis spp.; shooting, tilapia. MORTALIDAD DE AGUILAS PESCADORAS {PANDION HALIAETUS) INVERNANTES Y OTRAS AVES EN INFRAESTRUCTIRA PISCICOLA EN COLOMBIA Resumen. — El aguila pescadora {Pandion haliaetus) pasa el invierno en todo el territorio colombiano. La recaptura de aguilas pescadoras anilladas indican que muchas son cazadas en este pais con cifras en aumento desde 1970. La incidencia del numero de ^uilas muertas coincide con el desarrollo de la acuicultura en Colombia y la cria de tilapia roja {Oreochromis spp.). Esta industria tradicionalmente ha tenido problemas con aves depredadoras de peces por lo cual realizamos una encuesta en 83 granjas piscicolas en tres departamentos de Colombia en el 2001 para determinar las especies que consumen peces y el numero de aves eliminadas anualmente. Nuestros resultados mostraron que la depredacion por aves en la infraestructura piscicola ocurre en todo el territorio pero el problema es mas severo en la porcion sur del pais en el departamento del Huila. Los propietarios de las granjas reportaron que eliminan aguilas pescadoras en los tres departamentos, con pocos individuos en Antioquia (al norte de Colombia) equivalente a 5 individuos, y un numero maximo en el Huila de 270 individuos. Adicional- mente los propietarios reportaron que cazan otras nueve especies de aves las cuales incluyen a Chloloceryle americana, Pitangus sulphuratus, Egretta thula, Casmerodius albus, Phalacrocorax olivaceus, Nycticorax nycticorax, Ardea cocoi, Bubulcus ibis, Butorides striatus, para un total estimado de >9000 aves eliminadas anualmente en los tres departamentos, Los propietarios de las granjas reportaron que han probado otros metodos de control aparte de las armas de fuego, para reducir las perdidas tales como la instalacion de redes ^ E-mail address: mbechard@boisestate.edu ^ E-mail address: rapaces@audinet.com 292 December 2003 Osprey Mortality in Colombia 293 protectoras, cuerdas de alambre, espantapajaros, aparatos con sonidos, pero ninguno ha resultado efi- ciente in incluyendo el uso de las armas de fuego para la reduccion de las perdidas economicas causadas por aves. [Traduccion de los autores] Ospreys {Pandion haliaetus) breed throughout North America (Palmer 1988, Poole 1989, Johns- gard 1990, Poole et al. 2002). During the 1950s and 1960s, pesticide contamination threatened many populations and declines were widespread across the breeding range (Poole 1989). Since then, populations have rebounded and, today, the Osprey is again a common species in coastal, lake, and riverine habitats. The Osprey is migratory with populations from locations in the western U.S. win- tering in Mexico and Central America (Henny and Van Velzen 1972, Melquist et al. 1978, Johnson and Melquist 1991, Martell et al. 2001) and populations from coastal areas of the eastern U.S. and Canada and the Great Lakes region of the Midwest winter- ing in South America (Martell et al. 2001, Poole et al. 2002) . Ospreys banded as nestlings in Maryland, Virginia, New York, New Jersey, Michigan, and Wis- consin have mainly been recovered in Colombia, Venezuela, Equador, and Brazil indicating that northern South America is a primary wintering area for eastern and Midwestern populations (Henny and Van Velzen 1972, Poole and Agler 1987, Niemuth 1991). The Osprey is unusual because it is exclusively piscivorous eating a wide variety of fresh and salt- water fish species (Poole 1989, Poole et al. 2002). It has traditionally been called the “fish hawk” in North America and “aguila pescadora” in South America and it has been shot, had its nests robbed or torn down, and otherwise been persecuted in areas where people believe it competes with them for sport or commercial fisheries. The Osprey is protected from shooting and other forms of hu- man-caused mortality in North America but, on its Caribbean and Central and South vMnerican win- tering grounds, there are few laws restricting the killing of birds. Band recoveries of dead Ospreys indicate that shooting and trapping continue to threaten the species on its wintering grounds (Poole and Agler 1987, Santana and Temple 1987, Ewins and Houston 1992). The Osprey is also un- usual because it does not breed until three years of age and immatures remain on the wintering grounds continuously until they become sexually mature (Henny and Van Velzen 1972, Poole 1989). Therefore, shooting may selectively eliminate younger individuals and possibly decrease recruit- ment of new breeders into populations. Currently, there is insufficient information to judge the ex- tent and severity of the South American shooting threat, despite the fact that it may have increased since the 1970s (Poole and Agler 1987, Ewins and Houston 1992). Since the 1970s, fish farming or aquaculture has become a new and thriving business in Latin Amer- ica. Production systems consisting of extensive ranching operations where fry are released into reservoirs and later harvested as adults have been built in Mexico, Central and South America (Fitz- simmons 2000). Many aquaculture facilities spe- cialize in the production of tilapia {Oreochromis spp.), which are now produced in virtually every country of the Americas. By 1998, the annual pro- duction of tilapia had grown to 201 067 mt (metric tons) in Latin America and the U.S. imported 72 428 mt of live weight fish in 2000 (Fitzsimmons 2000). Aquaculture facilities typically raise tilapia in shallow ponds that are <1 ha in size, but, with facilities consisting of >30 ponds, they can have large areas of impounded water. Tilapia are sur- face-feeders and, when thousands of these bright red fish come to the surface to feed, they become easy prey for piscivorous birds. With its tremendous warm water resources, Co- lombia has become one of the leading aquaculture fish producers in South America (Fitzsimmons 2000). Commercial fish production began in the 1980s and by 1996, 22 states or departments were producing a mean of 25 063 mt of fish a year (Ins- tituto Nacional de Pesca y Aquicultura de Colom- bia [INPA] unpubl. data). Some commercial facil- ities raise rainbow trout {Oncorhynchus mykiss), but most are dedicated to the production of tilapia and Colombia produces a mean of 15 000 mt of tilapia each year (Fitzsimmons 2000) . There is an obvious potential for large numbers of piscivorous birds to be killed at these aquaculture facilities and, as such, these facilities may have some effect in redis- tributing populations of resident and wintering pi- scivorous birds in the country. Because of the es- calating numbers of Ospreys that are shot at 294 Bechard and Marquez-Reyes VoL. 37, No. 4 aquaculture facilities and brought to rehabilitation facilities in Colombia (pers. observ.) and increas- ing concern from the birding community about the numbers of Ospreys being killed each year (Nielson 1998), we undertook a survey to assess the extent of the threat to North American Ospreys wintering in Colombia from shooting, trapping, and other human-caused mortality at aquaculture facilities throughout the country. Study Areas and Methods We chose three departments, Antioquia in northern Colombia, Valle del Cauca in west-central Colombia, and Huila in southern Colombia, to conduct our survey. We chose these departments because they provided a good representation of the variation in the sizes and produc- tion of aquaculture facilities in the country, and each de- partment had ca. 30 commercial aquaculture facilities that were either licensed by the Corporacion de Valle del Cauca or by INPA. Antioquia (05°26'-08°52'N, 73°53'- 77°07'W) has a total of 28 licensed aquaculture facilities and was the largest of the three departments surveyed covering an area of 63 612 km^. The department of Valle del Canca (03°04'-05°02'N, 72°42'-74°27'W) encompass- es a 22 140 km^ area of west-central Colombia and has 32 licensed aquaculture facilities. The department of Huila in the southern portion of Colombia (01°33'- 03°47'N, 74°28'-76°36'W) is the smallest of the three de- partments covering an area of 19 890 km^ and it has 27 licensed aquaculture facilities. Of the 87 licensed aquaculture facilities in the three departments, we visited all except four to interview facil- ity owners and managers who were familiar with the daily operations of the facilities. The four facilities not visited were in the department of Valle del Cauca, and they were not surveyed either because the owners declined to par- ticipate or because they were in locations where condi- tions made them too dangerous to visit. During each visit, we administered a standard questionnaire to owners or managers who were familiar with the daily operations of facilities. Before administering the questionnaire, a state- ment signed by officials of INPA and the Ministry of the Environment was read stating that all answers would be kept confidential and that no legal proceedings would result from any answers given to the questioner. The questionnaire was administered verbally, and we com- pleted the answers on the questionnaire forms as the in- terviewees responded to them. The questions asked for information on the size of the facility (number of ponds and ha of impounded water), its annual fish production (mt) , if birds were a problem because they impacted an- nual fish production, the species of birds depredating fish, the seriousness of the impact by each species (on a scale from 1-5 with 1 being a species with one or a few individuals infrequently depredating ponds and 5 being a species with several individuals depredating ponds on a daily basis) , if birds were shot, which species of birds were shot, estimated number of each species shot each year (1-10, 11-20, 21-50, 51-75, 76-100, 101-200, or >200 shot annually), and alternative methods to shoot- ing that had been used to decrease the depredation Table 1. Questionnaire administered to aquaculture fa- cility owners and managers to estimate bird depredation and mortality at aquaculture facilities in the departments of Antioquia, Valle del Cauca, and Huila in Colombia. Facility name Permit number Location Name of water source Water temperature (°C) Number of ponds Area of impounded water (ha) Fish species cultivated Annual fish production (mt) Annual income (pesos) Do you have problems with bird depredation? How much do you estimate you lose annually to bird depredation? (pesos) Which species of birds are a problem and rank each species in terms of the seriousness of the problem (1 = none or little problem, 2 = slight problem, 3 = moderate problem, 4 = serious problem, 5 = severe problem with fish taken daily) . Do you shoot problem birds? If yes, estimate the num- bers of each species that are shot annually (1-10, 11- 20, 21-50, 51-75, 76-100, 101-200, >200). Have you used other methods of deterring birds from taking fish at your facility? If yes, please describe each method and rate its effectiveness. problem (Table 1). Most of the interviewees knew the local common names for the species of birds that caused depredation losses at their farms. When there was any question abont the identity of a species of bird, we used color photographs and color plates in Hilty and Brown (1986) to help interviewees identify the species. To esti- mate the numbers of each species shot annually at each facility, we used the midpoints of the ranges given by in- terviewees for the numbers of birds they shot each year. To rank species in terms of the seriousness of the threat they posed to the production of fish, we averaged the rankings given by the facility owners in each department. Results A total of 82 of the 83 aquaculture facilities sur- veyed reported experiencing depredation losses to the following 10 species of birds; Green Kingfisher (Chloroceryle americana), Great Kiskadee (Pitangus sulphuratus) , Snowy Egret {Egretta thula) , Great Egret {Casmerodius albus), Olivaceous Cormorant {Phalacrocomx olivaceus) , Black-crowned Night-Her- on {Nycticorax nycticorax) , White-necked Heron (Ar- dea cocoi), Cattle Egret {Bubulcus ibis), Striated Her- on {Butorides striatus), and Osprey (Table 2). Of the 82 facilities with depredation losses, 35 reported December 2003 Osprey Mortality in Colombia 295 Table 2. Mean ranking of birds that pose depredation problems at aquaculture facilities in the departments of Antioquia, Valle del Cauca, and Huila in Colombia and estimates of the number of each species shot annually. Antioquia Valle del Cauca Huila Species Ranking^ No. Shot RankincA No. Shot Ranking^ No. Shot Total Shot Green Kingfisher 2.6 44 2.5 65 3.5 1836 1945 Great Kiskadee 1.7 0 1.7 0 2.5 1692 1692 Snowy Egret 1.1 0 1.1 0 2.9 1621 1621 Great Egret 1.8 45 2.25 65 2.75 931 1041 Olivaceous Cormorant 1.8 230 1.4 19 1.0 288 537 Black-crowned Night-Heron 1.0 0 2.6 105 2.0 318 423 White-necked Heron 1.6 5 2.0 13 1.4 403 421 Cattle Egret 1.0 15 1.0 0 1.7 400 415 Striated Heron 1.3 0 2.5 35 1.1 364 399 Osprey 1.5 5 2.6 40 2.8 270 315 Total shot 356 342 8323 9021 ^ 1 = none or little problem, 2 = slight problem, 3 = moderate problem, 4 = serious problem, 5 = = severe problem with fish taken daily. shooting birds killing a combined estimate of >9000 birds per year. Facilities in all three depart- ments reported Green Kingfishers as causing the most serious losses and managers at these facilities reported killing more kingfishers (ca. 2000/yr) than any other species. Facilities also experienced depredation losses to Great Kiskadees, which took pelletized fish food and fingerlings, and managers at these facilities shot an estimated 1700 kiskadees per year. As a group, ciconiiforms were viewed as causing the most serious depredation losses at fish farms with Snowy Egrets, Great Egrets, and Black- crowned Night-Herons generally considered to be the biggest threats to facility production. Managers of facilities shot an estimated total of >3000 egrets and night-herons. An estimated 315 Ospreys were shot each year at facilities. Ospreys were consid- ered to cause the most serious depredation losses in southern Colombia in the department of Huila. The one facility that did not experience bird dep- redation problems was located in Antioquia and it raised primarily rainbow trout that were grown in completely-covered raceways and protected from birds. Aquaculture facilities in Antioquia ranged in size from 1-400 ponds {x = 36.8 ± 74.8, N = 28, ±SD) with 0.01-26.0 ha of impounded water (x = 1.8 ± 5.1, N = 28) raising a mean of 79.02 ± 155.9 mt of fish per year (range = 1.5-600, N = 28) . Twenty- two of the facilities raised mostly rainbow trout and the remaining six raised mainly tilapia. A total of 27 facilities in Antioquia experienced losses in pro- duction to bird depredation. At 23 facilities, birds ranked as causing the most serious depredation losses were Green Kingfishers, Great Egrets, Oli- vaceous Cormorants, Great Kiskadees, and White- necked Herons that took fish either while perching on overhead wires or while wading along the edges of ponds (Table 2). Only four facilities in Antio- quia ranked Ospreys as causing the most serious production losses, and all of these facilities spe- cialized in tilapia production. Eight facility man- agers said they shot an estimated 230 Olivaceous Cormorants, 45 Great Egrets, 44 Green Kingfish- ers, 15 Cattle Egrets, and 5 White-necked Herons annually, but only 5 Ospreys per year. Valle del Cauca, the second-largest, fish-produc- ing department in Colombia, had a mean annual fish production of 4560 mt, most of which was ti- lapia. The aquaculture facilities consisted of fewer ponds (x = 13.1 ± 13.3 ponds, range = 1-56, N — 28) but they were much larger in size (x = 5.5 ± 5.8 ha of impounded water, range — 0.4-18.2, N = 28), than in Antioquia. Twenty of the facilities raised a mean of 67.6 ± 90.6 mt (range = 5-240) of tilapia each year. All 28 of the facilities surveyed in Valle del Cauca reported losses in fish produc- tion to bird depredation. Black-crowned Night- Herons, Ospreys, Green Kingfishers, Striated Her- ons, and Great Egrets were considered to be problem species with mean rankings >2 (Table 2). Fourteen facility managers reported Ospreys to be a serious problem species. Only four facility man- agers in Valle del Cauca said they shot birds and 296 Bechard and MArquez-Reyes VoL. 3V, No. 4 estimated killing 105 Black-crowned Night-Herons, 65 Great Egrets, 50 Striated Herons, 40 Ospreys, and 20 White-necked Herons each year. The department of Huila was the smallest fish producer of the three departments surveyed, pro- ducing a mean of 2132 mt of fish per year, nearly 94% of which was tilapia. Aquaculture facilities in Huila had more ponds than in either Antioquia or Valle del Cauca {x — 62-1 ± 139,7 ponds, range = 3-650, N — 27), but they were smaller in size and covered a mean of 4.1 ± 6.1 ha (range = 0.2-25, N = 27) . Nevertheless, five of the facilities in Huila were very large consisting of >20 ponds and >15 ha of impounded water. Facilities in Huila were very productive, producing a mean of 99.08 ± 132.42 mt of tilapia (range = 2.5-840, N = 27) annually. In Huila, all 27 aquaculture facilities sur- veyed reported losses in fish production to bird depredation. Green Kingfishers, Great Egrets, and Great Kiskadees were again ranked as causing se- rious depredation problems (Table 2). Twenty-two facility managers reported Ospreys to be a serious depredation problem and 1 1 of these reported Os- preys to be their most serious problem species. Un- like the other departments, aquaculture facilities in Huila experienced serious depredation losses to Snowy and Cattle Egrets. Birds were shot at 23 fa- cilities with an estimated 8323 birds killed annually including 1836 Green Kingfishers, 1692 Great Kis- kadees, 1621 Snowy Egrets, 931 Great Egrets, 403 White-necked Herons, 400 Cattle Egrets, 364 Stri- ated Herons, 318 Black-crowned Night-Herons, and 270 Ospreys. One facility manager reported shooting >100 Ospreys at a communal roost tree located on the Magdalena River, where as many as 50 Ospreys would roost each night. The number he shot did not reduce the depredation problem at his facility because as many as 20 Ospreys fed there daily. The day we visited the facility, we ob- served 10 Ospreys taking fish from his ponds. Discussion Of the 22 departments in Colombia that have INPA-licensed commercial aquaculture facilities, annual fish production averages <100 mt in the departments of Arauca, Atlantico, Gasanare, Cau- ca, Cesar, Choco, Guaviare, and Vichada, and <1500 mt in the departments of Boyaca, Cundi- namarca, Risaralda, Putumayo, and Santander. Most of this fish production is rainbow trout so we did not consider facilities in any of these depart- ments to pose serious shooting threats to birds be- cause these facilities are relatively small and trout are mainly raised in covered raceways that are pro- tected from birds. The departments of Antioquia, Cordoba, Huila, Meta, Tolima, and Valle del Cauca each average >2000 mt of fish production annu- ally, ranging from a low of 2100 mt in Cordoba to a high of 6589 mt in Meta. These departments have the largest aquaculture facilities and raise mainly tilapia in large, shallow ponds that are prone to bird depredation. Therefore, these de- partments have the greatest potential for develop- ing bird depredation problems and the shooting of birds at aquaculture facilities. The results of our survey showed that several species of birds depredate aquaculture facilities in Colombia with the depredation problem apparent- ly increasing from north to south in the country. The most common solution to the problem that has been used by facility managers is shooting, and shooting appears to increase from as few as 100 birds shot in Antioquia in northern Colombia to as many as 8000 shot annually in Huila in southern Colombia. This increase appears to be due to a shift in the emphasis of fish production at aqua- culture facilities in the country from mostly trout production at facilities in northern Colombia to mostly tilapia production in southern Colombia. In northern Colombia, most facility managers do not consider the bird depredation problem to be se- vere enough to warrant killing birds and only eight of them said they shot kingfishers, herons, egrets, and Ospreys. In Huila, facility managers appear to see hird depredation as a more serious problem. This attitude was reflected in the numbers and va- riety of birds they shot. All of the fish farm man- agers who shot birds, also said that shooting was not an effective method of decreasing their losses because new birds simply replaced birds that were shot. The numbers of Ospreys reported shot each year also increased from five in Antioquia to over 270 in the southern department of Huila. Ospreys were considered to be only a minor problem in Antio- quia causing production losses at only those facil- ities that emphasized tilapia production. The shooting in Huila increased correspondingly with an increase in the productivity of tilapia in this de- partment. Warmer water temperatures in this de- partment are more conducive to the production of tilapia. Tilapia are raised in large, shallow im- poundments averaging nearly 5 ha in size. At any time, aquaculture facilities in Huila have as many December 2003 Osprey Mortality in Colombia 297 as 1 million tilapia in various stages of growth. The combination of the large amount of impounded water and the multitude of easily-captured fish prey are natural lures for Ospreys. Of the species that were shot, only Great Egrets, Snowy Egrets, Cattle Egrets, Black-crowned Night- Herons, and Ospreys are considered to be boreal migrants in northern South America (Davis 1993, Parsons amd Master 2000, McCrimmon et al. 2001, Ridgley and Greenfield 2001). All, excluding the Osprey, are also resident species that breed as well as winter in Colombia (Hilty and Brown 1986); therefore, it was impossible to assess the overall threat of Colombian aquaculture facilities to bo- real migrants. We recovered a total of six USGS bands that owners had removed from dead birds, but had not reported to the Bird Banding Labo- ratory. All were from dead Ospreys. Two were from Ospreys that had been banded as nestlings in Maine and one each came from Wisconsin, Virgin- ia, New York, and Connecticut, further supporting the view that Colombia is an important wintering area for Ospreys from the eastern and Midwestern U.S. Aquaculture of tilapia first began in the Ameri- cas in the 1960s and 1970s. Currently, Mexico is the biggest producer of tilapia but the industry is rapidly growing in Honduras, Costa Rica, Cuba, Ja- maica, Colombia, Ecuador, Peru, Venezuela, and Brazil (Fitzsimmons 2000). Future production trends include further intensification of produc- tion in every country with Brazil expected to be- come the higgest producer of tilapia in the next 20 yr (Fitzsimmons 2000) . Ospreys winter in all of these countries, therefore, if this trend continues, we could expect an increase in the shooting of Os- preys on their wintering grounds. In North Amer- ica, the construction of reservoirs for irrigation and flood control have had a substantial effect on the redistribution of breeding Ospreys, particularly in the western U.S. (Henny et al. 1978a, 1978b, Swenson 1981, Poole 1989). A similar phenome- non may be currently underway in Latin America. Aquaculture facilities may redistribute wintering populations of Ospreys away from natural habitats such as coastal, lake, and river areas to man-made reservoirs, where they are lured by the abundance of easily-captured fish prey. Unlike North America, where legislative controls and public education has reduced the risk of Ospreys being shot (Poole and Agler 1987), in Latin America shooting continues to be a very real threat. Many of the countries lack any laws that prevent shooting and those that do frequently lack enforcement. Shooting is an easy method of eliminating problem birds and band re- covery data from Central and South America sup- port that this is the usual method of reducing the avian depredation problem. Of the Ospreys band- ed in Canada, 39% have been recovered shot and the number shot appears to have increased since the 1970s, which coincides with the timing of the first construction of aquaculture facilities in Latin America (Poole and Agler 1987, Poole 1989, Ewins and Houston 1992). Populations of Ospreys con- tinue to increase throughout North America (Poole et al. 2002) indicating that the mortality at aquaculture facilities has probably not yet reached numbers great enough to impact populations of North American Ospreys. Nevertheless, in view of the current growing global trend in fish farming, mortality from shooting will only increase in years to come, perhaps developing into a serious prob- lem for Osprey populations in the eastern and Midwestern U.S. Our survey showed that fish farm owners had tried a variety of nonlethal bird deterrents to re- duce their annual losses to birds. The most widely- used alternatives to shooting were noise-making devices, such as guns, cannons, and fireworks, and patrols by people who flushed birds from the edges of ponds. None of these methods had proven ef- fective because problem birds apparently either be- came habituated to the noise makers or simply learned to ignore people patrolling ponds. A few facility managers had tried using scarecrows and dogs to frighten birds away. Neither of these meth- ods was effective after problem birds became ac- customed to them. Another widely-used deterrent was the installation of netting around ponds. Nets were frequently used to reduce loses to birds such as herons and egrets that hunt along the edges of ponds. Owners reported that the devices were rel- atively effective provided the netting was main- tained along the edges of ponds. Such mainte- nance was labor intensive and it was difficult to make certain that the netting was in place at all times. Only five fish farms reported having tried using overhead lines to deter depredation from ae- rial foragers such as Ospreys (Salomon and Conte 1981). The five that tried this method found it to be relatively inexpensive and potentially effective in reducing losses to large aerial-hunting birds like Ospreys. 298 Bechard and MArquez-Reyes VoL. 37, No. 4 Acknowledgments We thank E Ramirez, R Estela, and J.J. Orozco for help with surveys of fish farms. W. Weber, SAO, R Montoya, Corporacion de Valle del Cauca, and M.L. Vidal, Cor- poracion de Huila also provided support for the project. This study was funded by the Western Hemisphere Pro- gram, USFWS, Washington, DC, Agreement Number 14- 48-982 10-99-G41 6. Literature Cited Davis, W.E. 1993. Black-crowned Night-Heron (Nycticorax nycticorax) . In A. Poole and F. Gill [Eds.], The birds of North America, No. 74. The Academy of Natural Sciences, Philadelphia, PA, and The American Orni- thologists’ Union, Washington, DC U.S.A. Ewins, PJ AND C.S. Houston. 1992. Recovery patterns of Ospreys, Pandion haliaetus, banded in Canada up to 1989. Can. Field-Nat. 106:361-365. Fitzsimmons, K. 2000. Future trends of tilapia aquacul- ture in the Americas. Pages 252-264 in B.A. Costa- Pierce and J.E. Rakocy [Eds.], Tilapia aquaculture in the Americas, Vol. 2. World Aquaculture Society, Ba- ton Rouge, LA U.S.A. Henny, C.J. and W.T. Van Velzen. 1972. Migration pat- terns and wintering localities of American Ospreys. J. Wild! Manage. 36:1133-1141. , D.J. Dunaway, R.D. Mallelte, and J.R. Koplin. 1978a. Osprey distribution, abundance, and status in western North America: 1. The northern California population. Northwest Sci. 52:261-272. , J.A. Collins, and W.J. Deibert. 1978b. Osprey distribution, abundance, and status in western North America: II. The Oregon Population. Murrelet 59:1 4— 25. Hilty, S.L. and W.L. Brown. 1986. A guide to the birds of Colombia. Princeton Univ. Press, Princeton, NJ U.S.A. JOHNSGARD, P.A. 1990. Hawks, eagles, and falcons of North America. Smithsonian Institution Press, Wash- ington, DC U.S.A. Johnson, D.R. and W.E. Melquist. 1991. Wintering dis- tribution and dispersal of northern Idaho and eastern Washington Ospreys./. Field Ornithol. 62:517-520. Martell, M.S., C.J. Henny, P.E, Nye, and M.J. Solensky. 2001. Fall migration routes, timing, and wintering sites of North American Ospreys as determined by sat- ellite telemetry. Condor 103:715-724. McCrimmon, D.A., Jr., J.C. Ogden, and G.T. Bancroft. 2001. Great Egret {Casmerodius albus). In A. Poole and F. Gill [Eds.], The birds of North America, No. 570. The Academy of Natural Sciences, Philadelphia, PA, and The American Ornithologists’ Union, Washing- ton, DC U.S.A. Melquist, W.E,, D.R. Johnson, and W.D. Carrier. 1978 Migration patterns of northern Idaho and eastern Washington Ospreys. Bird-Banding 49:234—236. Nielson, G.J. 1998. Las aguilas pescadoras. Acuioriente 3: 10 . Niemuth, N.D. 1991. Recoveries of Osprey banded in Wisconsin. Passenger Pigeon 53:109-114. Palmer, R.S. 1988. Handbook of North American birds, Vol. 4. Yale Univ. Press, New Haven, CT U.S.A. Parsons, K.C. and T.L. Master. 2000. Snowy Egret {Egret- ta thula). In A. Poole and F. Gill [Eds.], The birds of North America, No. 489. The Academy of Natural Sci- ences, Philadelphia, PA, and The American Ornithol- ogists’ Union, Washington, DC U.S.A. Poole, A.F. 1989. Ospreys; a natural and unnatural his- tory. Cambridge Univ. Press, Cambridge, U.K. and B. Agler. 1987. Recoveries of Ospreys band- ed in the United States, 1914-1984. / Wildl. Manage 51:148-155. , R.O. Bierregaard, and M.S. Martell. 2002. Os- prey {Pandion haliaetus). In A. Poole and F. Gill [Eds.], The birds of North America, No. 683. The Birds of North America, Inc., Philadelphia, PA U.S.A. Ridgley, R.S. and P.J. Greenfield. 2001. The birds of Ec- uador. Cornell Univ. Press, Ithaca, NYU.S.A. Salomon, TP. and F.S. Conte. 1981. Control of bird damage at aquaculture facilities. Wildlife Manage- ment Leaflet No. 475, USFWS Cooperative Extension, Univ. California, Davis, CA U.S.A. Santana, E.C. and S.A. Temple. 1987. Recoveries of banded Ospreys in the West Indies. J. Field Ornithol 58:26-30. Swenson, J.E. 1981. Status of the Osprey in southwestern Montana before and after the construction of reser- voirs. West. Birds 12:47-51. Received 24 September 2002; accepted 29 August 2003 Associate Editor: Joan L. Morrison J. Raptor Res. 37(4):299-306 © 2003 The Raptor Research Foundation, Inc. BLOOD CHEMISTRY, CYTOLOGY, AND BODY CONDITION IN ADULT NORTHERN GOSHAWKS {ACCIPITER GENTILIS) Lauri a. Hanauska-Brown^ and Alfred M. Dufty, Jr. Raptor Research Center, Department of Biology, Boise State University, 1910 University Drive, Boise, ID 83725 US. A. GaryJ, Roloff Timberland Resources, Boise Cascade Corporation, 1564 Tomlinson Road, Mason, MI 48854 US. A. Abstract. — ^A bird’s physiological state contributes to its reproductive success and survival, yet few base- line physiological data have been published for wild raptors. Mean levels of protein, cholesterol, calcium, uric acid, aspartate aminotransferase, lactate dehydrogenase, and creatine kinase were measured in 29 Northern Goshawks {Accipiter gentilis) during 1998—99. None of these substances were significantly dif- ferent between males {N = 8) and females {N = 2) . Levels of all substances were highly variable among individual birds and unrelated to the body condition index (mass/wing chord X tail length X culmen length) . Total white blood cell count estimates and differential white blood cell counts were not signif- icantly different between the sexes. Of the blood cell measures, only the percent of heterophils and lymphocytes, and the ratio of heterophils to lymphocytes (H/L) differed between birds in good body condition and birds in relatively poor body condition. The IT/L ratio has recently been proposed as a reliable measure of stress. Thus, variable H/L ratios between groups of wild birds may indicate differ- ences in stress levels and overall health. Key Words: Northern Goshawk; Accipiter gentilis; blood chemistry; H/L ratio; body condition. QUIMICA SANGUINEA, CITOLOGIA, Y CONDICION CORPORAL EN ADULTOS DEL AZOR NOR- TENO {ACCIPITER GENTILIS) Resumen. — El estado fisiologico de una ave contribuye a su exito reproductivo y supervivencia, hasta el momenta muy pocos datos fisiologicos de linea base han sido publicados para rapaces silvestres. Los niveles medios de proteinas, colesterol, acido urico, aminotransferasa aspartato, lactato deshidrogenasa, y creatina quinasa se midieron en 29 azores nortenos {Accipiter gentilis) durante 1998—99. Ninguna de estas substancias fue significativamente diferentes entre machos {N = 8) y hembras {N = 2). Los niveles de todas las substancias fueron altamente variables entre aves individuales y no estuvieron relacionadas con el indice de condicion del cuerpo (masa/cuerda alar H longitud de la cola H longitud del culmen). El conteo de leucocitos totales estimados y el conteo diferencial de leucocitos, no fueron significativa- mente diferentes entre sexos. Entre las medidas de las celulas sanguineas, unicamente el porcentaje de heterofilos y linfocitos, y la razon de heterofilos a linfocitos (H/L) difirio entre aves con buena condi- cion corporal y aves en condicion corporal relativamente pobre. La razon H/L ha sido propuesta re- cientemente como una medida confiable de estres. Asi, las razones H/L variables entre grupos de aves silvestres pueden indicar diferencias en los niveles de estres y en la salud global. [Traduccion de Cesar Marquez] Northern Goshawks {Accipiter gentilis) are found throughout forested portions of the intermountain west (Squires and Reynolds 1997) where they are classified as a “sensitive species” by the United States Forest Service (Beals and Harris 1996, Squires and Reynolds 1997) and a “species of spe- ^ Present address: Idaho Dept, of Fish & Game, 1515 Lin- coln Rd., Idaho Falls, ID 83401 U.S.A.; e-mail address: lhanausk@idfg.state.id.us cial concern” by the Idaho Department of Fish and Game (Beals and Harris 1996). Basic knowl- edge of adult animal health, including blood- chemistry information, is needed for the man- agement of vulnerable species (Ferrer and Dobado-Berrios 1998), such as the goshawk. How- ever, past research on free-ranging goshawk life his- tory or management does not include an assess- ment of adult health from blood chemistry or cytology (blood cell) measures. These measures 299 300 Hanauska-Brown et al. VoL. 37, No. 4 provide information regarding nutritional and im- munological status, and help us understand eco- logical and behavioral problems (Ferrer 1990, Stein et al. 1998). Two objectives were addressed in this study: (1) determine baseline means and ranges of blood chemistry and cytology levels in adult breeding goshawks, and (2) determine the relationships among blood chemistry, blood cytol- ogy, and body condition. Blood chemistry is an indirect method of assess- ing health as blood acts as a means to mobilize and transport nutrients, metabolic products, immune cells, and hormones (Brown 1996). Metabolic sub- stances present in blood plasma reflect avian nu- tritional health and overall condition (Snyder and Terry 1986, Mauro 1987, Ferrer 1993). Much like plasma metabolites, plasma enzymes also reflect avian health. Enzymes such as aspartate aminotransferase (AST), lactate dehydrogenase (LDH), and creatine kinase (CK) help regulate metabolic reactions and respond to changes in the body (Mauro 1987). Interpretation of plasma en- zyme levels is complicated by natural variation un- less levels are significantly higher or lower than ref- erence samples (Ivins et al. 1986, Campbell 1995). Differential white blood cell counts provide in- formation on the activity level of a bird’s immune system and stress response (Dhabhar et al. 1996). Heterophils and lymphocytes are the predominant white blood cells found in raptors (Lavin et al. 1992, Stein et al. 1998). The ratio of heterophils to lymphocytes (H/L) has proven to be a reliable measure of stress in poultry and wild birds and a more accurate measure of stress than individual cell numbers or corticosterone levels (Gross and Siegel 1983, Maxwell 1993, Vleck et al. 2000). Finally, morphological characteristics measured to provide a body condition estimate can be indic- ative of individual health (Brown 1996). Body mass IS relative to structural size (i.e., wing, tail, culmen length), which is defined as the nutrient-reserve independent size of an individual (Piersma and Davidson 1991). Birds in good condition should have a relatively high ratio of mass to structural size (Brown 1996). We hypothesized that goshawks as- sessed to be in good body condition by this index would exhibit relatively high levels of blood chem- istry measures (e.g., protein, calcium) indicative of adequate nutrition and low levels of cytology mea- sures (e.g., H/L ratio) indicative of an activated immune system, high stress, or disease. Materials and Methods This study was conducted from 1998-99 in the west- central mountains of Idaho. The study area encompassed parts of the Boise, Payette, Snake, and Salmon River drainages where elevations range from 460-3050 m (Steele et al. 1981). Public and private lands were sur- veyed for goshawks in Washington, Adams, Idaho, Valley, and Boise counties. The study area is a mix of lowland meadows and rangeland and higher elevation mixed co- niferous forests. A more complete description of the veg- etation and climatic conditions can be found in Hanaus- ka-Brown (2000). We trapped adult goshawks with a live Great Horned Owl {Bubo virginianus) lure to elicit aggressive responses from nesting goshawks and capture them in a dho-gaza net (Detrich and Woodbridge 1994). Upon capturing one adult from a nest, we reset the dho-gaza net in an attempt to catch the other adult. We collected up to 500 pi of whole blood after punc- turing the brachial vein with a sterile lancet. Plasma was sent to IDEXX Veterinary Services Laboratory, Inc. (Port- land, OR) which measured protein, cholesterol, calcium, uric acid, and three blood enzymes: AST, CK, and LDH. Additional substances were measured in fewer than 10 individuals: albumin, globulin, phosphorous, potassium, sodium, amylase, and alkaline phosphatase. Descriptive statistics only are presented for these variables as statisti- cal analyses were deemed inappropriate. We evaluated adjusted Sidak P-values (Zar 1996) from multiple pairwise comparisons (MULTTEST, SAS Release 6.12, 1996) to determine significant differences between samples collected from females and males, with N ^ 8 samples for each sex (Zar 1996). Adjusted Sidak P-values are slightly less conservative than the Bonferroni test and control for comparison-wise error (Zar 1996). We per- formed Spearman correlation analyses of metabolic sub- stances and enzymes when N ^ 8 individuals for each sex. Assumptions of normality and heterogeneity were tested prior to statistical comparisons, and correlations and nonparametric tests were performed when assump- tions were violated. A blood smear was made for each bird and sent to IDEXX Veterinary Services Laboratory, Inc. The labora- tory estimated total white blood cell numbers and per- centages of cell types including: heterophils, eosinophils, monocytes, and lymphocytes. We used multiple pairwise comparisons and Sidak P-values again to evaluate mean differences between the sexes. We also performed Spear- man correlation analyses for all cytological measures. Eor the calculation of body condition, we measured mass, unflattened wing chord length, tail length, and cul- men length. Wing chord length was measured to the tip of the longest primary, tail length to the tip of the longest rectrix, and culmen length the distance between bill tip and anterior end of the cere. The mass:structural size ratio was determined by dividing mass (g) by the product of wing chord length (mm), tail length (mm), and cul- men length (mm). Preliminary results from our data demonstrated blood metabolite and enzyme measures (N = 7) were signifi- cantly correlated, as were white blood cell measures {N = 6) . Thus, we predicted that a suite of blood chemistry and cytology measures, rather than any individual mea- December 2003 Goshawk Blood and Body Measures 301 Table 1. Metabolites and metabolic enzymes measured in Northern Goshawks in west-central Idaho during the breeding seasons of 1998 and 1999 along with ranges reported for 24 captive adult female and male goshawks (Benyon et al. 1996). Substance Females Males Benyon et ai.. (1996) N Mean (SE) Range N Mean (SE) Range Range Total protein (g/dL) 21 1.8 (0.2) 0.9-3.4 8 2.3 (0.2) 1. 3-3.2 2.6-4.2 Cholesterol (mg/dL) 21 155.1 (14.8) 33.0-286.0 8 201.0 125.0-282.0 154.7-444 7 Uric acid (mg/dL) 21 12.6 (2.3) 1.8-40.2 8 20.9 (6.6) 9.4-63.0 8.6-14.4 Calcium (mg/dL) 21 4.7 (0.5) 1.8-9. 5 8 5.9 (0.7) 3. 3-8.0 8.6-10.9 Phosphorous (mg/dL) 4 3.7 (0.7) 2.4-5.0 4 2.2 (0.5) 1.2-3. 7 NA Potassium (mEq/L) 5 3.0 (0.3) 2.5-4.0 4 2.5 (0.3) 1.9-3. 1 NA Sodium (mEq/L) 5 158.6 (7.4) 136.0-181.0 4 159.8 (2.3) 155.0-165.0 NA A/G ratio^ 4 0.7 (0.1) 0.5-1.0 4 0.7 (0.03) 0.6-0.7 NA Globulin (g/dL) 4 1.6 (0.2) 1.2-2.2 4 1.6 (0.1) L4-1.8 NA Albumin (g/dL) 4 1.1 (0.1) 0.9-1 .2 4 1.0 (0.1) 0.9-1. 2 NA AST (lU/L)^’ 21 286.2 (26.2) 108.0-561.0 8 375.5 199.0-556.0 176.0-409.0 LDH (lU/L)" 21 347.6 (47.9) 136.0-840.0 8 616.4 322.0-1040.0 120.0-906 0 CK (IU/L)‘i 21 127.0 (12.3) 56.0-285.0 8 160.6 56.0-412.0 218.0-775 0 Amylase (lU/L) 4 900.5 (44.1) 790.0-1006.0 4 864.0 567.0-1195.0 NA Aik. phos.'" (lU/L) 4 20.5 (4.7) 9.0-31.0 4 20.5 (4.8) 15.0-35.0 NA ^A/G = albumin/globulin ratio. ^ AST = aminotransferase. ^ LDH = 1 acetate dehydrogenase. CK = creatine kinase. Aik. phos. = alkaline phosphatase. sure, would best represent goshawk physiological health. We hypothesized that goshawks with higher body condi- tion indices would exhibit a more healthy physiological prohle compared to birds in relatively poor body condi- tion. To test this hypothesis we reduced the physiological data from each bird to an individual factor score using principle components analysis (StatSoft 1998). The factor score was a multi-dimensional physiological representa- tion of goshawk health. In this analysis, we used the factor scores associated with the first eigenvector. Subsequently, factor scores for the 10 goshawks in the best and the worst body condition (as indicated by body condition in- dex) were evaluated using a nonparametric two-sample comparison test (Mann-Whitney G-test; Bailey 1995). We used the body condition index as the measure of overall condition as it was independent of the other physiologi- cal measures (Hanauska-Brown 2000) and has been used to represent health in other studies (Rising and Somers 1989, Brown 1996). We combined sexes in the analysis because there were no differences in blood chemistry, cytology, or body condition means. Two-sample compar- isons were performed on factor scores derived from the entire suite of physiology variables {N = 13), the com- bined metabolic and enzyme variables {N = 7), and the suite of blood cell variables {N = 6) . For significant two- sample comparisons, factor loadings from the principle components analysis were used to help determine which variables had greatest influence on the factor scores (StatSoft 1998). Factor loadings can be interpreted as the correlations between factors (i.e., the set of factor scores for all birds) and the individual physiological variables (StatSoft 1998). Thus, the magnitude and sign of factor loadings can be used to determine the relative impor- tance of individual variables in explaining the two-sample comparison results. Results Metabolite, enzyme, and white blood cell con- centrations were measured in 21 adult female and 8 adult male goshawks (Tables 1-2). There were no between-sex differences in metabolite or en- zyme levels (P > 0.50). Data on metabolic sub- stances and enzymes from 24 captive adult gos- hawks (Benyon et al. 1996) are provided for comparison with our data (Tables 1-2). There were no between-sex differences in total white blood cell estimate, individual cell type, or H/L ratio for goshawks measured in this study {P > 0.45); however, we note that the differences be- tween female and male monocyte numbers ap- proached significance {t = —3.10, P = 0.07). The correlation analysis of all blood chemistry variables showed that all metabolites and enzymes were cor- related (P ^ 0.05) with the exception of the en- zyme CK. Correlation analysis of the blood cell 302 Hanauska-Brown et al. VoL. 37, No. 4 Table 2. White blood cell total estimates (XlOVl) and white blood cell types (%) measured in Northern Goshawks in west-central Idaho along with ranges reported for 43 captive adult female and male goshawks (Benyon et al. 1996). WBC Females Males Benyon et al. (1996) N Mean (SE) Range N Mean (SE) Range Range Total estimate 21 9.8 (1.0) 4.0-20.0 8 8.7 (1.2) 4.0-13.0 4.0-11.0 Monocytes 21 2.3 (0.4) 0.0-6.0 8 5.3 (1.3) 0.0-12.0 1.0 Lymphocytes 21 38.2 (2.5) 16.0-58.0 8 46.8 (5.0) 26.0-66.0 14.0-19.0 Eosinophils 21 19,6 (1.4) 6.0-33.0 8 13.6 (2.3) 4.0-23.0 7.0 Heterophils 21 40.0 (2.4) 19.0-58.0 8 34.3 (5.0) 12.0-54.0 36.0-65.0 H/L ratio‘s 21 1.27 (0.17) 0.37-3.31 8 0.89 (0.23) 0.18-2.00 NA ^ Heterophils to lymphocytes. counts and white blood cell estimates demonstrat- ed significant patterns {P ^ 0.05) among several variables. The mean body condition ratio was 4.39 ± 0.08 (range = 3.81-4.96) for females and 4.90 ± 0.03 (range = 3.86-6.16) for males (Table 3). There was no significant difference between fe- male and male body condition ratios (Wilcoxon signed rank, z = 1.51, P = 0.13). The two-sample comparison test using all physi- ological variables {N = 13) for the 10 goshawks with the highest and lowest body condition scores detected no difference {N = 20, U— 27, P= 0.25). Similarly, the comparison using the combined me- euzymc variables {N — 7) showed no difference between goshawks of different body condition {N = 20, U = 36, P — 0.29). Two-sample comparison tests on factor scores derived from blood cell variables {N = 6), however, showed a difference between good and poor body condition goshawks {N — 20, U = 3, P < 0.01). Factor load- ings for each variable indicated that the H/L ratio (negative loading) was the variable primarily re- sponsible for the observed difference. These re- sults are consistent with paired comparisons per- formed on individual variables that demonstrated high lymphocyte numbers, low heterophil num- bers, and low H/L ratios for goshawks in good con- dition (Hanauska-Brown 2000) . There was a signif- icant negative correlation between H/L and body condition {N — 30, = —0.48, P < 0.01). Discussion Concerns over the population status of goshawks throughout the intermountain west region (Ken- nedy 1997) prompted us to seek a means of index- ing the health of adult breeding goshawks. We based this index on morphological measurements, blood chemistry, and cytological characteristics. We focused on these variables because other research- ers have questioned the assessment of population viability using occurrence, density, or annual pro- ductivity estimates (e.g., Van Horne 1983, Franklin et al. 2000) . Our data represent baseline metabolite, enzyme, and blood cytology levels for a sample of free-rang- ing goshawks in central Idaho. The only other baseline data available for goshawks are from “nor- mal, healthy” male and female goshawks of various ages, housed in rehabilitation centers (Benyon et al. 1996). Captive birds are provided with a contin- uous food source, but also experience a multitude of stressors (e.g., confinement and handling) not Table 3. Morphology measures from Northern Goshawks in west-central Idaho. Morphology Females Males N Mean (SE) Range N Mean (SE) Range Mass (g) 21 1006.8 (14.8) 870-1134 8 766.9 (15.7) 710-824 Wmg (mm) 21 353.8 (1.7) 341-372 8 325.4 (3.8) 310-337 Tail (mm) 21 261.7 (1.2) 249-272 8 225.1 (2.4) 212-232 Culmen (mm) 21 24.8 (0.3) 23.2-29.9 8 21.1 (0.3) 19.7-25.1 December 2003 Goshawk Blood and Body Measures 303 present in the wild (Stein et al. 1998). Captive birds receive a steady, high protein diet (Gee et al. 1981, Garcia-Rodriguez et al. 1987) that provides an adequate supply of amino acids for the synthesis of blood proteins (Ferrer et al. 1987). Indeed, total protein was higher for the captive goshawks than for the birds sampled in this study. Wild goshawks may experience periods of fasting (Newton 1979) and may be forced to specialize on certain prey species (Newton 1979, Younk 1996) leading to a mineral or dietary deficiency (Snyder and Terry 1986). Birds in the wild engage in activities (e.g., courtship, territoriality, nestling care) that reduce the time available for foraging, which also may lead to dietary deficiencies, particularly in poor-quality habitat. Calcium levels were lower in wild than in captive goshawks (Benyon et al. 1996), which also can be attributed to dietary differences. Captive birds re- ceiving chicken carcasses, rabbits, and other ‘whole’ food items (Gee et al. 1981) obtain calcium from ingesting bones and other body parts. Con- versely, captive birds fed only red meat (including muscle, heart, and liver) were found to be defi- cient in calcium (Graham and Halliwell 1986). Di- etary calcium in the wild comes from a varied diet including whole prey items (Graham and Halliwell 1986, Squires and Reynolds 1997). However, wild birds may not be able to capture a steady diet of suitable prey, and low levels of calcium may reflect fasting periods (Halliwell 1981). The mean uric acid level for females was within the range reported for captive goshawks, but the mean for males was above the reported range (Benyon et al. 1996) by more than 6 mg/dL. Uric acid, a nitrogenous waste product, is typically ele- vated in birds with high protein diets (Bell and Sturkie 1965, Gee et al. 1981). High uric acid levels in free-ranging birds can also be attributed to de- creased body condition or food stress (Handrich et al. 1993, Balbontm and Ferrer 2002, Casado et al. 2002). Males are under high energy demands during the breeding season as they hunt for the adult female and young, as well as for themselves. Mean cholesterol levels from sampled birds were within the range of cholesterol values reported for captive goshawks (Benyon et al. 1996). The three enzymes we measured are quite vari- able in captive goshawks (Benyon et al. 1996). Such individual differences are common in avian species due to natural variation in habitat, genetics, body condition, environmental influences, and other factors (Gee et al. 1981, Hoffman et al. 1985, Stein et al. 1998). For example, Stein et al. (1998) documented more variability in wild populations of American Kestrels {Falco sparverius) and Red-tailed Hawks {Buteo jamaicensis) than in captive birds. Mean levels of AST and LDH measured in this study were within the wide range of enzyme levels in captive goshawks (Benyon et al. 1996). However, lower levels of CK were measured in wild versus captive birds (Table 1). Average or low CK levels can indicate enhanced muscle strength and endur- ance (Apple and McGue 1983, Knuth and Chaplin 1994). For example, flight training in captive Red- tailed Hawks improved the structural integrity of muscles and lowered plasma CK levels (Knuth and Chaplin 1994). The low levels of CK observed in this study may reflect increased muscle strength and flight endurance (Apple and McGue 1983, Knuth and Chaplin 1994) in the wild birds. Individual variation due to proximate factors may play a major role in variable blood chemistry measures, but many other influences can affect the measurement of health in wild birds. Garcia-Ro- driguez et al. (1987) demonstrated diurnal and cir- cadian rhythms of calcium, uric acid, and choles- terol levels in raptors. We could not control the time of day of sample collection because of the logistical constraints of trapping breeding gos- hawks. Time of sample collection in this study ranged from 0700-1900 H. Variation in the levels of protein, uric acid, calcium, and cholesterol in individual goshawks in this study, therefore, may be attributed to recent feeding, fasting periods, or cir- cadian and diurnal rhythms. Geographic location also may affect blood chem- istry measures. Soil types and environmental con- tamination affect physiology, as raptors incorpo- rate locally-obtained nutrients and toxins from their prey items (Hoffman et al. 1985, Ferrer and Dobado-Berrios 1998). Differences in total protein, uric acid, cholesterol, and calcium were found be- tween two populations of Adalbert’s Eagle {Aqmla adalberti) foraging in different regions of Spain (Ferrer and Dobado-Berrios 1998). Qualitative dif- ferences in each population’s diet were suggested as the cause of the physiological differences (Fer- rer and Dobado-Berrios 1998). Similarly, differenc- es in prey abundance, prey availability, and habitat in central Idaho may be reflected in goshawk blood chemistry measures. For example, goshawks in the western region of the study area likely con- sumed more ground squirrels {Spermophilus spp.) 304 Hanauska-Brown et al. VoL. 37, No. 4 than goshawks in the northern regions, which probably consumed more birds (pers. observ.). Goshawk samples from this study showed in- creased numbers of lymphocytes, monocytes, and eosinophils compared to the ranges reported for captive goshawks (Benyon et al. 1996). Increased levels of particular leukocytes or white blood cell total estimates can indicate physiological stress or decreased immunocompetence (Smith and Bush 1978, Kontecka et al. 1999). A shift from normal values in any direction can indicate disease (Camp- bell and Dein 1984). Blood loss or tissue damage at the collection site can also affect white blood cell counts. Our blood collection method may have overestimated circulating white blood cells because hemorrhage induced by puncturing the wing vein produces tissue damage that triggers clotting and attracts white blood cells. Capillary blood in hu- mans (e.g., blood collected into capillary tubes af- ter finger pricks) has higher white blood cell counts than blood collected intravenously (Daee et al. 1988, Kayiran et al. 2003). We minimized this potential change in white blood cells by collecting blood immediately upon puncturing the vein. The H/L ratio was significantly different be- tween birds in relatively good, compared to birds in relatively poor, condition. Such differences sug- gest decreased immunocompetence or higher rates of stress in birds of poor condition. The H/ L ratio has been shown to be a less variable indi- cator of stress than differential counts and more reliable than corticosterone levels in reflecting chronic stress (Vleck 2002). One of the main objectives of this work was to explore the relationships between a noninvasive body condition index and the blood chemistry and cytology of wild breeding goshawks. We expected less variability in the blood parameters we observed and stronger relationships among the various phys- iological measurements and body condition than we detected. Our small sample size undoubtedly influenced variability in our measures, and some of the assumptions underlying the mass/length technique we used to assess body condition have recently been questioned (Green 2001). Further- more, considering that all birds sampled were breeding, our “poor” condition birds may not have been in poor condition relative to the overall wild population. Despite these concerns, our hnd- ing that the relative differences in body condition among breeding birds were reflected in the H/L ratio highlights the potential sensitivity and utility of this technique. Our results also illustrate the difficulty in evalu- ating the physiological state of a bird at any given time. Metabolite and enzyme levels appear to be influenced by too many external and individual factors to serve as accurate barometers of individ- ual health. Despite the limitations inherent in physiological studies of wild raptors such as ours, the H/L ratio and body condition index hold promise as indicators of adult goshawk health. Fur- ther study of free-living raptor physiology is war- ranted to assess the value of the H/L ratio as a measure of individual health. Acknowledgments We thank the following organizations for funding and supporting this project: Boise Cascade Corporation, Boi- se State University, and the Idaho Department of Fish and Game. We thank the biologists of the Boise and Pa- yette National Forests for their cooperation and time. Fi- nally, we thank field technicians Trent Brown and Greg Burak, and graduate student Lynda Leppert. Literature Cited Apple, F.S. and M.K. McGue. 1983. Serum enzyme changes during marathon training. Am. J. Clin. Pathol. 79:716-719. Bailey, N.T.J. 1995. Statistical Methods in Biology, 3rd Ed. Cambridge Univ. Press, Cambridge, U.K. Bal bontin, J. and M. Ferrer. 2002. Plasma chemistry ref- erence values in free-living Bonelli’s Eagle {Hieraaetus fasdatus) nestlings./. Raptor Res. 36:231-235. Beals, J.W. and C.E. Harris. 1996. Northern Goshawk nest area monitoring in the Twin Falls Ranger District of the Sawtooth National Forest. Unpubl. Report, Ida- ho Fish and Game, Boise, ID U.S.A. Bell, D.J. and P.D. Sturkie. 1965. Chemical constituents of blood. Pages 32-84 in P.D. Sturkie [Ed.], Avian Physiology, 2nd Ed. Comstock Publ. Assoc., Ithaca, NY U.S.A. Benyon, P.H., N.A. Forbes, and N.H. Harcourt-Brown. 1996. Manual of raptors, pigeons, and waterfowl. Brit- ish Small Animal Vet. Assoc. Ltd., Iowa State Univ Press, Ames, lA U.S.A. Brown, M.E. 1996. Assessing body conditions in birds Pages 67-135 in V. Nolan, Jr. and E.D. Ketterson [Eds.], Current Ornithology, Vol. 13. Plenum Press, New York, NY U.S.A. Campbell, T.W. 1995. Avian hematology and cytology Iowa State Press, Ames, lA U.S.A. AND F.J. Dein. 1984. Avian hematology: the basics. Vet. Clin. North Am. Small Anim. Pract. 14:223-248. Casado, E., J. Balbontin, and M. Ferrer. 2002. Plasma chemistry in Booted Eagle {Hieraaetus pennatus) dur- December 2003 Goshawk Blood and Body Measures 305 ing breeding season. Comp. Biochem. Physiol. 131:233- 241. Daee, L.N., S. Halvorsen, P.M. Mathisen, and K. Mi- RONSKA. 1988. A comparison between haematological parameters in ‘capillary’ and venous blood from healthy adults. Scand. J. Clin. Invest. 48:723-726. Detrich, P.J. AND B. WOODBRIDGE. 1994. Territory fidel- ity, mate fidelity, and movements of color-marked Northern Goshawks {Accipiter gentilis) in the southern Cascades of California. Stud. Avian Biol. 16:130-132. Dhabhar, F.S., A.H. Miller, B.S. McEwen, and R.L. Spen- cer. 1996. Stress-induced changes in blood leukocyte distribution: role of adrenal steroid hormones. Im- munology 157:1638—1644. Ferrer, M. 1990. Hematological studies in birds. Condor 92:1085-1086. . 1993. Blood chemistry studies in birds: some ap- plications to ecological problems. Trends Comp. Bio- chem. Physiol. 1:1031-1044. AND P. Dobado-Berrios. 1998. Factors affecting plasma chemistry values of the Spanish Imperial Ea- gle, Aquila adalberti. Comp. Biochem. Physiol. 120:209— 217. , T. Garcia-Rodriguez, J.C. Carrillo, andJ. Cas- TROViEjo. 1987. Hematocrit and blood chemistry val- ues in captive raptors (Gyps fulvus, Buteo buteo, Milvus migrans, Aquila heliaca). Comp. Biochem. Physiol. 87: 1126-1127. Franklin, A.B., D.R. Anderson, R.J. Gutierrez, and K.P. Burnham. 2000. Climate, habitat quality, and fitness in northern Spotted Owl populations in northwestern California. Ecol. Monogr. 70:539-590. Garcia-Rodriguez, T., M. Feriier, F. Recia, and J. Cas- TROVIEJO. 1987. Circadian rhythms of determined blood chemistry values in buzzards and eagle-owls. Comp. Biochem. Physiol. 88:663-669. Gee, G.F., J.W. Carpenter, and G.L. Hensler. 1981. Spe- cies differences in hematological values of captive cranes, geese, raptors, and quail./. Wild! Manage. 45: 463-483. Graham, D.L. and W.H. Halliwell. 1986. Malnutrition in birds of prey. Pages 379-385 in M.E. Fowler [Ed.], Zoo and Wild Animal Medicine. W.B. Saunders Co., Philadelphia, PA U.S.A. Green, A. 2001. Mass/length residuals: measures of body condition or generators of spurious results? Ecology 82: 1473-1483. Gross, W.B. and H.S. Siegel. 1983. Evaluation of the het- erophil/lymphocyte ratio as a measure of stress in chickens. Avian Dis. 27:972-979. Haijjwell, W.H. 1981. Serum chemistry profiles in the health and disease of birds of prey. Pages 111-112 in J.E. Cooper and A.G. Greenwood [Eds.], Recent Ad- vances in the Study of Raptor Diseases. Chiron Pubk, West Yorkshire, U.K. Hanauska-Brown, L.A. 2000. Evaluation of Northern Goshawk {Accipiter gentilis) productivity, nestling growth, and adult health in relation to a central Idaho habitat potential model. M.S. thesis, Boise State Uni- versity, Boise, ID U.S.A. Handrich, Y, L. Nicolas, and Y. Le Meho. 1993. Winter starvation in captive common Barn Owls; physiologi- cal state and reversible limits. Auk 110:458-469. Hoffman, D J-. J .C. Franson, O.H. Pattee, G.M. Bunck, AND H.C. Murray. 1985. Biochemical and hematolog- ical effects of lead ingestion in nestling American Kes- trels {Falco sparverius). Comp. Biochem. Physiol. 80:431- 439. Ivins, G.K., G.D. Weddle, and W.H. Halliwell. 1986 Hematology and serum chemistry in birds of prey. Pages 280—290 in M.E. Fowler [Ed.], Zoo and Wild Animal Medicine. W.B. Saunders Co., Philadelphia, PA U.S.A. Kayiran, S.M., N. Ozbek, M. Turan, and B. Gurakan 2003. Significant differences between capillary and ve- nous complete blood counts in the neonatal period. Clin. Lab. Haematol. 25:9—16. Kennedy, P.L. 1997. The Northern Goshawk {Accipiter gen- tilis) : is there evidence of a population decline? J. Rap- tor Res. 31:95—106. Knuth, S.T. and S.B. Chaplin. 1994. The effect of exer- cise on plasma activities of lactate dehydrogenase and creatine kinase in Red-tailed Hawks. L Raptor Res. 28: 27-33. Kontecka, H., J.M. Ksiazkiewicz, and L. Nogowski. 1999. Effects of different stressors on laying rate and selected blood indices in reproductive ducks. Amm Feed Sci. Technol. 8:63-72. Lavin, S., R. Cuenca, I. Marco, R. Velarde, and L. Vina 1992. Hematology and blood chemistry of the Marsh Harrier {Circus aeruginosus) . Comp. Biochem. Physiol 103:493-495. Mauro, L. 1987. Hematology and blood chemistry. Pages 269-275 in B.A. Giron Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor Management Techniques Manual National Wildlife Federation Sci- ence and Technical Series, No. 10. Port City Press, Inc., Baltimore, MD U.S.A. Maxwell, M.H. 1993. Avian blood leukocyte responses to stress. World’s Poult. Sci. J. 49:34-43. Newton, I. 1979. Population Ecology of Raptors. T. & A.D. Poyser, London, U.K. Piersma, T. and N.C. Davidson. 1991. Confusions of mass and size. Auk 108:441-444. Rising, J.D. and K.M. Somers. 1989. The measurement of overall body size in birds. Auk 106:666-674. SAS Release 6.12. 1996. SAS User’s Guide. SAS System Inst. Inc., Carey, NC U.S.A. Smith, E.E. and M. Bush. 1978. Haematologic parame- ters on various species of strigiformes and falconifor- mes. Wildl. Dis. 14:447-450. Snyder, R.L. and J. Terry. 1986. Avian nutrition. Pages 190-200 in M.E. Fowler [Ed.], Zoo and Wild Animal Medicine. W.B. Saunders Co., Philadelphia, PAU.S A 306 Hanauska-Bro^^tv et al. VoL. 37, No. 4 Squires, J.R. and R.T. Reynolds. 1997. Northern Gos- hawk {Accipiter gentilis) in A. Poole and F. Gill [Eds.], The Birds of North America, No. 298. Academy of Natural Sciences, Philadelphia, PA, and American Or- nithologists’ Union, Washington, DC U.S.A. StatSoft, Inc. 1998. STATISTIGA for Windows, Com- puter program manual. StatSoft, Inc., Tulsa, OK U.S.A. Steele, R., R.D. Phister, R.A. Ryker, and J.A. Kittams. 1981. Forest Habitat Types of Central Idaho. USDA Forest Service, Gen. Tech. Rep. INT-114, Ogden, UT U.S.A. Stein, R.W., J.T. Yamamoto, D.M. Fry, and B.W. Wilson. 1998. Comparative hematology and plasma biochem- istry of Red-tailed Hawks and American Kestrel win- tering in California./. Raptor Res. 32:163-169. Van Horne, B. 1983. Density as a misleading indicator of habitat quality. /. Wildl. Manage. 47:893-901. VleCK, C.M. 2002. Comparison of corticosterone and het- erophil to lymphocyte ratios as indicators of stress in free-living birds. Pages 401-411 in A. Dawson and C.M. Chaturvedi [Eds.], Avian Endocrinology. Narosa Publishing House, New Delhi, India. Vleck, C.M., N. Vertalino, D. Vleck, and T.L. Bucher. 2000. Stress, corticosterone, and heterophil to lym- phocyte ratio in free-living Adelie Penguins. Condor 102:392-400. Younk, J.V. 1996. Breeding ecology of the Northern Gos- hawk in relation to surface gold mining in naturally fragmented aspen forests of northern Nevada. M.S. thesis, Boise State University, Boise, ID U.S.A. Zar, J.H. 1996. Biostatistical Analysis, 3rd Ed. Prentice Hall, Upper Saddle River, NJ U.S.A. Received 9 July 2002; accepted 28 July 2003 J. Raptor Res. 37(4) :307— 314 © 2003 The Raptor Research Foundation, Inc. PRODUCTIVITY AND POPULATION TRENDS OF OSPREYS IN THE KAWARTHA LAKES REGION, ONTARIO, 1978-2001 Shane R. de Solla and Pamela A. Martin^ Canadian Wildlife Service, Environment Canada, Box 5050, 867 Lakeshore Road, Burlington ON, L7R 4A6 Canada Peter J. Ewins World Wildlife Fund, 245 Eglington Avenue E., Toronto, ON M4P 3J1 Canada Michael E. Barker Wetland Habitat Fund, 1289 Algonquin Blvd. Peterborough, ON K9H 6N1 Canada Abstract. — Osprey {Pandion haliaetus) populations were monitored extensively in the Kawartha Lakes region of Ontario, Canada, from 1978-2000, and intensively monitored in a subsample of these lakes, particularly Sturgeon Lake, (1991-2001). The number of occupied nests within the Kawartha Lakes region increased from 18 nests in 1978 to 89 nests in 1992 following the ban of organochlorine pesticides and the provision of artificial nesting structures. Surveys in 1996 and 2000, however, revealed 78 and 66 occupied nests, respectively, indicating a recent population decline or stabilization. The Osprey breeding colony at Sturgeon Lake followed very similar trends in nest numbers, peaking in the mid- 1990s, followed by a decline and apparent population stabilization by 2001. The mean annual produc- tivity in this colony between 1986 and 2001 was 1.17 chicks per occupied nest, and exceeded the re- ported replacement level of 0.8 chicks per nest. Population fluctuations and trends in overall annual production of fledglings were similar between the Sturgeon Lake breeding colony and birds breeding at lower densities in neighboring lakes. Estimated potential recruitment to the Sturgeon Lake colony, based on the number of fledglings produced 3-5 years previously, was positively related to population size in a given year {E = 0.65); but very few banded nestlings returned to the colony as breeders. Factors affecting productivity and population trends seem to operate similarly at local and regional scales. As very few fledged juveniles returned to the Sturgeon Lake colony as breeders, yet the population was maintained, we suggest there was probably considerable dispersal of juveniles among local natal sites throughout the region. Key Words: Osprey, Pandion haliaetus; Ontario; population; productivity. PRODUCTIVIDAD Y TENDENCIAS POBLACIONALES DE AGUILAS PESGADORAS EN LA REGION DE LOS LAGOS KAWARTHA, ONTARIO, 1978—2001 ReSUMEN. — Las poblaciones de Aguila pescadora (Pandion haliaetus) fueron monitoreadas extensiva- mente en la region de los lagos Kawartha de Ontario, Canada, desde 1978-2000, y monitoreadas inten- sivamente en una submuestra de estos lagos, partieularmente en el Lago Esturion, (1991—2001). El numero de nidos ocupados dentro de la region de los lagos Kawartha incremento desde 18 nidos en 1978 a 89 nidos en 1992 luego de la prohibicion de los pesticidas organoclorados y de proveer estruc- turas artificiales para anidacion. Los estudios en 1996 y 2000 sin embargo, revelaron 78 y 66 nidos ocupados, respectivamente, indicando un reciente declive en la poblacion o su estabilizacion. La colonia de anidacion del aguila pescadora en el Lago Esturion siguio tendencias muy similares en el numero de nidos, alcanzando su pico a mediados de los 90s, seguido por un declive y una aparente estabilizacion de la poblacion para el 2001. la productividad media anual en esta colonia entre 1986 y 2001 fue 1.17 polluelos por nido ocupado, y excedio el nivel de reposicion reportado de 0.8 polluelos por nido. Las fluctuaciones de la poblacion y las tendencias en la produccion anual global de juveniles fueron similares entre la colonia de anidacion del Lago Esturion y las aves que anidaron en mas bajas densidades en lagos vecinos. El reclutamiento potencial estimado para la colonia del Lago Esturion, basada en los numeros de juveniles producidos 3-5 ahos atras, se relaciono positivaraente con el tamano de la pobla- ^ Corresponding author’s e-mail address; pamela.martin@ec.gc.ca 307 308 DE SOLLA ET AL. VoL. 37, No. 4 cion en un ano dado = 0.65); aunque, muy pocos juveniles anillados retornaron a la colonia como reproductores. Los factores que afectan la productividad y las tendencias poblacionales parecen operar similarmente a escalas locales y regionales. A pesar que muy pocos de los juveniles levantados retornaron a la colonia del Lago Esturion como reproductores, aun la poblacion se mantuvo, sugerimos que alii probablemente bubo una considerable dispersion de juveniles entre los sitios natales locales de toda la region. [Traduccion de Cesar Marquez] Osprey {Pandion haliaetus) populations, like those of many fish-eating birds throughout North America, declined due to pesticide-related effects in the 1950s and 1960s, subsequently recovering following restrictions in the use of organochlorine pesticides and other chlorinated hydrocarbons in the early 1970s (Spitzer et al. 1983, Ewins et al. 1995, Houghton and Rymon 1997). Many pro- grams have been implemented to monitor Osprey populations in North America and to improve re- cruitment (Spitzer et al. 1983, Steidl et al. 1991, Castellanos and Ortega-Rubio 1995, Ewins 1996, Witt 1996). In 1978, the Ontario Ministry of Natural Re- sources began monitoring the Osprey population of the Kawartha Lakes region of south-central On- tario within the lower Great Lakes basin. Once common throughout the area, this population suf- fered a decline during the previous decades and the main objectives at that time were to assess and monitor population trends, and to enhance pop- ulation growth and reproductive success through the creation of artificial nesting sites (Barker 1986, 1988). In 1991, the Canadian Wildlife Service ini- tiated a study to assess the suitability of Ospreys as indicators of chlorinated hydrocarbon contami- nants within the Great Lakes basin, as well as to determine if Ospreys breeding on the Great Lakes were continuing to suffer any adverse effects of these contaminants (Ewins et al. 1995, Martin et al. 2003). Ospreys on several lakes within the Kawar- tha Lakes system were also monitored to provide comparative data on what was assumed to be a clean inland breeding population. However, PCB concentrations in the eggs and plasma of nestling Ospreys in Sturgeon Lake were found to be con- siderably higher than those of birds nesting on the Canadian Great Lakes and in the Kawartha Lakes region (Martin et al. 2003). By 1996, when it had been demonstrated that contaminant levels were low elsewhere in the study area, the Canadian Wildlife Service ceased Osprey monitoring activi- ties in the Great Lakes basin except in the Stur- geon Lake colony, where monitoring was contin- ued until 2001. We report here the population growth and fluc- tuations of Ospreys in the Kawartha Lakes region from 1978-2001 including data from survey re- ports from Ontario Ministry of Natural Resources from 1978-91 with those collected by the Canadian Wildlife Service and Kawartha Lakes Eriends of the Osprey Society from 1991-2001. As well, we ex- amine post-recovery patterns of productivity and adult returns of Ospreys in a selected small breed- ing colony. Study Area and Methods The study area was primarily within the Kawartha Lakes region in south-central Ontario, a transitional zone bordering the Canadian Shield and the Great Lakes-St. Lawrence Lowlands ecozones (Fig. 1 ) . This area consists of a chain of 17 shallow, mainly eu trophic lakes intercon- nected by a series of rivers and man-made canals. The water levels are controlled in some areas through the use of dams, and the littoral regions of some lakes were flooded as a result of the construction of the Trent-Sev- ern Waterway over the last two centuries. Although the total area is approximately 6600 km^, much of this land is under agricultural use and unsuitable Osprey habitat. The entire Kawartha Lakes region, as well as adjoining portions of the Trent-Severn Waterway including Scugog Lake, the Otonabee River, and Rice Lake, were searched for Osprey nests by Ontario Ministry of Natural Resourc- es in 1978-80, 1985, 1986, 1988, and 1991, and by the Canadian Wildlife Service in 1992. Suitable nesting hab- itat was searched in early June along routes with known nest sites, both by boat and by vehicle on roads near the lakes. Nest checks were done with a spotting scope or an extendible pole with an attached mirror. Second visits in July were made to nests throughout the region in 1986, 1988, and 1991; we calculated productivity on a per nest basis for these years (M. Barker unpubl. data). In 1988, 1991, 1996, and 2000, the entire region was also surveyed by fixed-wing aircraft from which nests may be viewed easily (Fuller and Mosher 1987), during the nestling pe- riod (midjune). Furthermore, we had access to nest site information collected by local Osprey enthusiasts. Thus, we believe that the majority of occupied Osprey nests in the Kawartha Lakes region were located during these sur- veys. A subset of Kawartha Lakes at the northwest end of the chain, including Sturgeon, Balsam, Canal, and Mitchell lakes, and Emily Creek (44°36'N, 7°20'W to 44°36'N, 79°05'W; Fig. 1) were surveyed annually by boat from December 2003 Trends in Ontario Ospreys 309 Figure 1. Kawartha Lakes study area indicating its position in the Great Lakes basin and the Sturgeon Lake Osprey breeding colony. 1991-96. The south end of Sturgeon Lake near the town of Lindsay was the site of a fairly dense nesting “colony” of Ospreys, in which up to 17 occupied nests occurred within a 3.5 km radius, In 1991, PCB contaminant levels in this colony were determined to be unusually high (egg X = 7.1 ppm; range: 2.02-26.5 ppm wet weight), and thus intensive annual monitoring in this colony continued un- til 2001. Ospreys in Sturgeon Lake and the other northwestern lakes nested mostly on flooded stumps, whose tops were within a meter of the water, and on man-made elevated platforms, although some nests were located on live or dead trees and utility poles (P. Martin unpubl. data). All known nests on each lake were visited twice a year by boat. The first visit was made in late May or early June, during mid- to late incubation, and was a check for nest occupancy. We did not visit during the laying period or early incubation to avoid causing nest abandonment. Thus, it was possible some nests were destroyed before our initial visit, but this probability remained constant among years and probably did not represent a bias in monitoring. The second visit was in late June or early July when the chicks were 5-6 wk old (banding age). At this time we banded nestlings with both a US Department of Interior issued aluminum numeric band and a single col- ored plastic band. Each color of plastic band indicated a yearly cohort, and was used by all Osprey banding pro- jects throughout the Great Lakes. The number of chicks remaining in the nest at this time were considered to be fledged although it is possible that some mortality may have occurred between banding and nest departure. The occupied nest (ON) was used as the census unit, and was defined as a nest attended by a pair of Ospreys, regardless of any breeding attempt. Therefore, an occu- pied nest was considered equivalent to a pair of potential breeders. A successful nest was one in which at least one young was reared to banding age. Productivity rate was defined as the number of chicks reared to banding age per occupied nest. Production was the total number of chicks fledged in the overall population or group (Steen- hof and Kochert 1982). We calculated the adjusted re- cruitment productivity (ARP) , which is an estimate of the potential for new breeders in a given year (Spitzer 1980, Spitzer et al. 1983). Spitzer et al. (1983) calculated that 50% of Ospreys first bred at 3, 30% at 4, and 20% at 5 yr of age. Thus, the number of potential breeders re- cruited into the population in a given year was weighted by these proportions for the number of fledglings 3, 4, and 5 yr prior, using the equation: [ 1 ] (0.5 X X_3 + 0.3 X Yt„4 + 0.2 X Yt_5), where Y is the number of fledglings, and t is the year To convert recruitment to productivity, the weighted sum of fledglings was divided by the number of occupied nests the year before their recruitment, Ot_i. From 1992-96, most adults (all female) Ospreys at Sturgeon Lake were trapped using a dome-shaped noose carpet (Bloom 1987) tied over the nests during late in- cubation when nest-site tenacity was strong and chances of nest abandonment were low. The banding operation took less than 15 min and birds were always observed to return to their nest. Banded adults were of unknown age. Each year during the initial nest check, an effort was made to read the large 2-digit bands of breeding birds with 8 X 42 binoculars or a spotting scope. Birds were not recaptured on the nest for the purpose of determin- ing their identity. Data Analysis. We estimated changes in the size of the 310 DE SOLLA ET AL. VOL. 37, No. 4 n •S’ I ! S' ■p 90 ■ • . 1 - ol 80 70 • ' • 60 50 40 30 “ : i ' i ’ 0 ■ B 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 Year Figure 2. Productivity, number of occupied nests, and total number of Osprey fledged 1978-2001 in the entire Kawartha Lakes region, Ontario. 2.4 2.0 1.6 1.2 j 0.8 I 1 0.4 32 28 24 8,20 1 16 S = '5, Cm ll I 8 4 1 1 "X ' pi •v:- 2.4 2.0 1.6 1.2 I 0.8 0.4 0.0 1985 1987 1989 1991 1993 1995 1997 1999 2001 Year Figure 3. Productivity, number of occupied nests, and total number of Osprey fledged 1985—2001 at the Stur- geon Lake colony, Ontario. Osprey population of the entire Kawartha Lakes region study area from 1978 onward, and in the Sturgeon Lake colony from 1986-2001, using least squares regression. We evaluated the proportion of fledglings from Stur- geon Lake and the four neighboring lakes that later re- turned to breed. We included only juveniles that were banded from 1991—98, as juveniles fledged after 1998 would not be expected to breed until at least 2002. We compared recruitment and ARP with the annual percent change in population size using least squares regression. Only 6 yr of data (1996-2001) were available for analysis, as a minimum of 5 yr prior to the given year were re- quired for the estimation of ARP. Data were insufficient for other areas within the Kawartha Lakes region to con- duct this analysis. We calculated the encounter rates of adults in Sturgeon Lake from 1994-2001. The identities of breeding birds at each nest were tabulated from 1994 onward. Adults were identified by band numbers where possible, or were listed as either unbanded or unknown (band unreadable or unable to determine presence of band) . The presence of plastic color bands was recorded. Weather data were collected by Environment Canada at the Peterborough Airport from 1994-2001 and were accessed on the Ontarioweather webpage (Patrick 2002). For the months of May and June, we calculated the means for the daily mean, maximum, and minimum tem- peratures, windspeed, and dewpoint. We examined the relationships between spring weather variables and Os- prey population dynamics at Sturgeon Lake. Results Kawartha Lakes Region: Population Size. From 1978-92 there was a four-fold increase in the num- ber of occupied nests throughout the Kawartha Lakes region study area (Fig. 2; = 0.997, P < 0 . 001 ), representing a mean annual rate of in- crease of 10.3%; this rate of increase was greatest between 1980 and 1985 (15.9%). However, the population had declined 9% by 1996 and 22% by 2000, relative to maximum nest numbers of 1992. Sturgeon Lake: Population Size. The number of occupied nests within the Sturgeon Lake colony increased from 11 in 1986 to 17 in both 1992 and 1993 (r2 = 0.91, ^ = 21.37, P = 0.04), fluctuated through 1996, then declined steadily to a low of nine in 2000 , although there was a slight resur- gence of nests in 2001 (r^ = 0.69, F, 7 = 15.31, P = 0.006; Fig. 3). Sturgeon Lake: Productivity. Mean annual pro- ductivity at the Sturgeon Lake colony ranged from 0.8-2.18, vdth a mean of 1.17 fledglings per occu- pied nest (Fig. 3). There was no relationship be- tween the total number of fledglings produced in the colony and the number of occupied nests in the same year, from 1986-2001 (F = 0.05, P = 0.45); however, nest success was significandy relat- ed to mean annual per nest productivity (r^ = 0.72, P< 0.001). The percent change in population size was positively correlated to ARP (F = 0.71, Fj 4 = 9.93, P = 0.035) . Similarly, there was a positive re- lationship between potential recruitment and the population size (r^ = 0.65, Fj 4 = 7.38, P = 0.053). Detailed population and productivity data were available for Ospreys nesting at much lower den- sities in four adjacent lakes within the northwest- ern portion of the Kawartha Lakes (Balsam, Canal, and Mitchell lakes, and Emily Creek) , providing a local comparison to the colony at Sturgeon Lake. The total number of occupied nests from these lakes ranged from 8-20 nests (Fig. 4). The popu- lation fluctuations observed in the Sturgeon Lake colony between 1986 and 2000 were similar to those in nearby lakes ( 1 ^ = 0.44, P = 0.052) and both groups reached their lowest numbers in 2000. The annual changes in the total number of fledg- lings produced per year was similar between the December 2003 Trends in Ontario Ospreys 311 Year Figure 4. Productivity, number of occupied nests, and total number of Osprey fledged 1985-2001 at Balsam, Canal, and Mitchell lakes, and Emily Creek, Ontario. Yr Observred in Subsequent Yeais Figure 3. Cumulative proportion of banded adult Os- preys returning per year from year of banding. Sturgeon Lake colony and the neighboring lakes (r^ = 0.51, Fi j — 7.28, P = 0.03; however, mean annual productivity per nest was not correlated be- tween the two areas (r^ = 0.34; P — 0.17). Never- theless, mean annual productivity did not differ be- tween the two groups pooled across years (dependent ^-test, ^ = 1.73, P = 0.13). Resightmgs of Banded Juveniles. In the north- western portion of the Kawartha Lakes a total of 179 chicks were banded between 1991 and 1998, and were thus available for recruitment into the breeding population by 2001, our last year of study. Only one bird banded as a juvenile was confirmed breeding at Sturgeon Lake; a total of three are known to have returned to the Kawartha Lakes re- gion. As well, in 1997, a male Osprey that had been banded with a plastic color band as a juvenile in 1993 bred successfully at Sturgeon Lake. However, the natal location of this bird is unclear, as in 1993, color banding of annual cohorts was being con- ducted throughout the Great Lakes basin. Band re- coveries were obtained for three other birds band- ed as juveniles. Two were found dead in their first autumn in Ohio and Michigan. The third was found dead on its wintering grounds in Lesser An- tilles during its second winter. Returns of Banded Adults. Of 18 adults banded at the nest in the Sturgeon Lake colony between 1992 and 1996, 15 were females. Two males re- turned to breed at the colony three years subse- quent to banding, while one was only observed in the year of banding. Of the females, two bred at the colony only the year of banding and were not observed in subsequent years. There were pro- nounced declines in the proportion of adults re- turning on years 2 and 6 (Fig. 5) . Females bred at the colony a modal frequency of 2 yr (six individ- uals) and a mean of 3.44 (SD = 2.09) years; seven was the maximum number of years females bred at the colony (two individuals). Thus, the modal minimum age at last breeding was 5 yr and the maximum age was 10 yr. Females with a high pro- portion of successful nests tended to have many breeding attempts (r^ = 0.64, “ 8.35, P = 0.014). However, females were equally likely to be successful in the final year observed breeding (65.4%) than any other year of breeding (64.3%; < 0.005, P = 0.97). There was an equal proba- bility that successful and unsuccessful females would switch nest sites the subsequent year (x^ = 2.23, P= 0.14). Effects of Weather. During the years 1994-2001, we found significant relationships between mean productivity of Ospreys in the Sturgeon Lake col- ony and weather conditions for May, but not for June. There was a significant positive relationship between productivity and mean daily temperature (r^ = 0.63, “ 10.41, P — 0.018), mean maxi- mum daily temperature (r^ = 0.48, F-^ q = 5.50, P — 0.058), mean minimum daily temperature (r® = 0.65, Fi 6 = 11.30, P = 0.015) and mean dewpoint (r^ = 0.67, Pi 6 ~ 12.02, P = 0.013). There were no significant relationships between productivity and mean, maximum, and gust wind speed. Discussion Osprey populations throughout North America have rebounded from the declines of the mid-cen- tury induced by organochlorine pesticide expo- sure, following the ban on the use of these chem- icals (Postupalsky 1989, Ewins et al. 1995, Henny and Kaiser 1996, Houghton and Rymon 1997). Al- though the Osprey population of the Kawartha 312 DE SOLLA ET AL. VOL. 37, No. 4 Lakes region in the lower Great Lakes basin has also expanded by four-fold from 1978-92, its status pre-DDT era is unknown. By the year 2000, how- ever, this regional population had contracted to 74.2% of its maximum expansion achieved in 1992. Similar leveling off or slight decline of Osprey pop- ulations, following periods of rapid growth, has oc- curred elsewhere in the Great Lakes basin. Osprey populations increased dramatically during the post-DDT era in Michigan (Postupalsky 1989); however, during the latter half of the 1990s, pop- ulations in the Upper and Lower Peninsula de- clined by 22% and 9%, respectively (S. Postupalsky pers. comm.). An Oregon population experienced similar increases in size throughout the 1970s and 1980s as contaminant levels declined (Henny and Kaiser 1996). In contrast, instead of reaching a pla- teau, this population more than doubled from 1995-2001 (USGS 2002). The authors suspect that recent acceptance by the Ospreys of utility poles and channel markers as nesting sites accounts for this sudden population increase (USGS 2002), sug- gesting that appropriate nest sites were limiting previously. Sturgeon Lake appeared to have a sur- plus of established nesting sites. Given that there were always some unoccupied nest platforms, and that we frequently observed the construction of new nests on exposed stumps, it seems that avail- ability of nest sites probably was not limiting this population of Ospreys. A common strategy in en- couraging Osprey population expansion has been to erect artificial nest sites (Ewins 1996, Witt 1996). Our results suggest that even at high population densities, nest site availability was not a limiting fac- tor. The Sturgeon Lake breeding colony seems to have reached a plateau and a management strategy based on further construction of artificial nests is unlikely to result in further growth. Nevertheless, there is anecdotal evidence (R. Skitch, Ontario Power Generation pers. comm.) that in the eastern section of the Kawartha Lakes region (Rice and Stoney lakes), there has been a dramatic increase in the use of utility poles in 2001 and 2002. This tendency to use utility poles may have facilitated an increase in the regional population overall. Although the eggs of Sturgeon Lake Ospreys were found to contain relatively elevated egg PCB concentrations when examined in 1991-92, there was no apparent correlation between productivity with either DDE or PCB concentrations through- out the Canadian Great Lakes basin, including the Kawartha Lakes region (Martin et al. 2003). Un- fortunately, there is neither contaminant nor pop- ulation data available for Canadian Great Lakes Osprey from the 1960s and 1970s with which to document a causal relationship. Nevertheless, it is unlikely that current levels of contamination ad- versely impact reproduction in the Sturgeon Lake colony (Poole 1989). The proportion of nests that were successful as opposed to the total number of nests occupied in a given year was strongly correlated to mean an- nual productivity, suggesting that the factors con- tributing to nest failure also affected partial mor- tality within a nest. Depredation was clearly evident for three complete clutches of eggs, whereas four complete broods of chicks were obviously depre- dated. Although only three Osprey nestlings were actually found dead in the nest at time of banding that were suspected to have died from starvation, there were many instances of partial brood reduc- tion (in either the egg or nestling stage). In some cases, single unhatched eggs remained in the nest at banding, possibly as a result of infertility or in- clement weather during incubation. The latter possibility is supported by the positive relationship between May temperatures and productivity. Flood- ing caused by storms, as well as controlled water level changes as a result of damming, are suspected to be the cause of destruction of nest contents in low, overwater stump nests (P. Martin, M. Barker, and B. Puxley unpubl. data). Other studies have also indicated that poor weather conditions im- pacted fledging success of Ospreys (Poole 1982, Machmer and Ydenberg 1990) but see Stinson (1978) and Stinson et al. (1987). Trends in annual production of successful fledglings were similar be- tween Sturgeon Lake and the neighboring lakes, suggesting that regulatory factors (e.g., weather) operate at a scale that encompasses the entire Ka- wartha Lakes region in a given year. Estimates of the productivity required to main- tain stable populations for Ospreys ranged from 0.8 fledglings per nest (Spitzer et al. 1983) to be- tween 1.22 and 1.30 (Henny and Ogden 1970). An important distinction between these studies (Spit- zer et al. 1983, Henny and Ogden 1970) was that the populations studied had very different trajec- tories of growth. The population studied by Spitzer (1980) was recovering, while those examined by Henny and Ogden (1970) were generally declin- ing. Nevertheless, 0.8 fledglings per nest is typically accepted as the productivity “break-even” point for Ospreys in the northeastern United States and December 2003 Trends in Ontario Ospreys 313 Canada (Twins et al. 1995). This rate was met or exceeded in each of the 11 years of study at the Sturgeon Lake colony and in all but one year in the neighboring lakes. Adjusted recruitment pro- ductivity (ARP) , which was based on fledgling pro- duction from 3-5 yr prior to any given year, was highly correlated with population size. Thus, re- duced productivity from 1995 through 1997 di- rectly correspond to declines in the breeding pop- ulation from 1998 through 2000. Also, the sudden increase in productivity in 1998 corresponded to the increased breeding population of 2001. The productivity “break-even” point does not account for differences in adult survivorship or dis- persal. Although we banded adult birds we were unable to estimate longevity as the ages of the birds were unknown at time of banding. Of 15 adult fe- males banded, two marked in 1992 were observed to breed on Sturgeon Lake for seven seasons. As- suming they were first time breeders when initially banded in 1992 and were 3 yr old at time of band- ing, these birds were a minimum of 10 yr old dur- ing their last known breeding year of 1998. Of 17 known-aged females who reached breeding age banded as nestlings in a recovering population in Michigan during the 1970s and 1980s, five reached or exceeded 10 yr of age whereas 12 died prior to attaining 10 yr of age (Postupalsky 1989). Thus, our rate of 13% of females exceeding 10 yr of age is lower than that of 35% in the Michigan birds, suggesting possible problems in adult survival, al- though data are limited. Less than 2% of 179 fledglings that were banded in the northwestern portion of the Kawartha Lakes region from 1991-98 were later encountered breeding in the Kawartha region. After 1996, how- ever, only the Sturgeon Lake colony was monitored intensively, so opportunities to observe banded birds were limited. Nevertheless, even at Sturgeon Lake, where 70% of all juveniles produced were banded over the 11 yr period, less than 3% of the 94 birds returned to this colony as breeders. This suggests that new breeders at the Sturgeon Lake colony arrived via dispersal from other natal sites, either in the Kawartha Lakes region or from more distant populations. The distances among the Ka- wartha Lakes are certainly within typical dispersal distances of Ospreys (Spitzer et al. 1983, Postu- palsky 1989). That the adjusted recruitment pro- ductivity at Sturgeon Lake was highly correlated with population size, despite the fact that very few juveniles produced there actually return to the col- ony to breed, suggests that trends in localized groups of breeders are indicative of regional pop- ulation fluctuations. The growth of the ICawartha Lakes population of Ospreys has stabilized or slightly declined in the last decade, but as rates of productivity seem to be adequate for reproduction, we do not expect fur- ther declines. Factors affecting productivity and population trends appear to operate similarly at the local and regional scale. As very few fledged juveniles returned to the Sturgeon Lake colony as breeders, yet the population was maintained, we suggest there was probably considerable dispersal of juveniles among local natal sites throughout the region. Although we did not detect nest site limi- tation at the Sturgeon Lake colony, it is possible that recent observed increases in the usage of util- ity poles in another portion of the Kawartha Lakes region may have fostered population increases throughout the region. Acknowledgments Funding for much of this work was supplied by The Ontario Ministry of Natural Resources and Environment Canada’s Great Lakes Action Plan. M. Bacro, L. Benner, and G. Barrett provided assistance in the field. We are grateful for the Kawartha Lakes Friends of the Osprey Society for logistical support and for maintaining the ar- tificial Osprey nest platforms. We would like to thank D. Patrick for the free access to the archived weather data from the Peterborough area. We would also like to thank the Otonabee Region Conservation Authority and the former Ontario Ffydro and the present Ontario Power Generation for their cooperation. This manuscript was improved by comments from D. Weseloh, C. Ffenny, and two anonymous reviewers. Literature Cited Barker, M, 1986, Osprey nesting survey. Lindsay District. Unpublished report submitted to Ontario Ministry of Natural Resources. Lindsay, ON, Canada. . 1988. Osprey nesting survey. Lindsay District. Un- published report submitted to Ontario Ministry of Natural Resources. Lindsay, ON, Canada. Bloom, P.H. 1987. Capturing and handling raptors. Pag- es 99-123 in B.A. Giron Pendleton, B.A. Millsap, K.W. Gline, and D.M. Bird [Eds.], Raptor management techniques manual, Natl. Wild!. Fed., Washington, DC U.S.A. Castellanos, A. and A. Ortega-Rubio. 1995. Artificial nesting sites and Ospreys at Ojo de Liebre and Guer- rero Negro lagoons, Baja California Sur, Mexico. J Field Ornithol 66:117-127. Ewins, PJ- 1996. The use of artificial nest sites by an in- creasing population of Ospreys in the Canadian Great Lakes Basin. Pages 109-123 in D. Bird, D.E. Varland, 314 DE SOLLA ET AL. VOL. 37, No. 4 and JJ. Negro [Eds.], Raptors in human landscapes. Academic Press, London, U.K. , S. POSTUPALSKY, T. WeISE, AND E.M. ADDISON. 1995. Changes in the status, distribution and biology of Ospreys {Pandion haliaetus) breeding on Lake Hu- ron. Pages 273-290 in M. Munawar, T. Edsall, and J. Leach [Eds.], The Lake Huron ecosystem: ecology, fisheries, and management. SPD Academic Publish- ing, Amsterdam, The Netherlands. Fuller, M.R. and J.A. Mosher. 1987. Raptor survey tech- niques. Pages 37-65 in B.A. Giron Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor management techniques manual, Natl. Wildl. Fed., Washington, DC U.S.A. Henny, C.J. and J.C. Ogden. 1970. Estimated status of Osprey populations in the United States. J. Wildl. Manage. 34:214-217. andJ.L. Kaiser. 1996. Opsrey population increase along the Willamette River, Oregon, and the role of utility structures, 1976-1993. Pages 97-108 mD. Bird, D.E. Varland, andJ.J. Negro [Eds.], Raptors in human landscapes. Academic Press, London, U.K. Houghton, L.M. and L.M. Rymon. 1997. Nesting distri- bution and population status of U.S. Ospreys 1994. J. Raptor Res. 31:44-53. Machmer, M.M. and R.C. Wjenberg. 1990. Weather and Osprey foraging energetics. Can. J. Zool. 68:40-43. Martin, P.A., S.R. de Solla, and P.J. Ewins. 2003. Chlo- rinated hydrocarbon contamination in Osprey eggs and nestlings from the Canadian Great Lakes Basin, 1991-1995. Ecotoxicology 12:209-224. Patrick, D. 2002. Ontarioweather webpage, http:// www.ontarioweather.com. Poole, A.F. 1982. Brood reduction in temperate and sub- tropical Ospreys. 53:111-119. — — ^ — . 1989. Ospreys: a natural and unnatural history. Cambridge Univ. Press, Cambridge, U.K. and New York, NYU.S.A. PosTUPALSKY, S. 1989. Osprey. Pages 297-313 in 1. New- ton [Ed.], Lifetime reproduction in birds. Academic Press, London, U.K. Spitzer, P.R. 1980. Dynamics of a discrete coastal breed- ing population of Ospreys in the northeastern United States during a period of decline and recovery, 1978. Ph.D. dissertation, Cornell University, Ithaca, NY U.S.A. , A.E Poole, and M. Schieibel. 1983. Initial pop- ulation recovery of breeding Ospreys in the region between New York City and Boston. Pages 231-241 in D.M. Bird [Ed.], Biology and management of Bald Eagles and Ospreys. Harpell Press, Ste. Anne de Belle- vue, Quebec, Canada. Steenhof, K. and M.N. Kochert. 1982. An evaluation of methods used to estimate raptor nesting success. J. Wildl. Manage. 46:885-893. Steidl, R.J., C.R. Griffin, and L.J. Niles. 1991. Contam- inant levels of Osprey eggs and prey reflect regional differences in reproductive success. J. Wildl. Manage. 55:601-608. Stinson, C.H. 1978. The influence of environmental con- ditions on aspects of the time budgets of breeding Ospreys. Oecologia 36:127-139. , J. Lauthner, and R.T. Ray. 1987. The effect of weather conditions on the behavior of Ospreys in northwestern Washington. Can.]. Zool. 65:2116-2118. USGS. 2002. Ospreys in Oregon and the Pacific North- west. USGS FS-1 53-02, U.S. Department of Interior, U.S. Geological Survey, Corvallis, OR U.S.A. Witt, J.W. 1996. Long-term population monitoring of Osprey along the Umpqua River in western Oregon. J. Raptor Res. 30:62—69. Received 20 December 2002; accepted 17 June 2003 Associate Editor: Marco Restani /. Raptor Res. 37(4) :315-322 © 2003 The Raptor Research Foundation, Inc. HABITAT CHARACTERISTICS OF GREAT GRAY OWL SITES IN THE SISKIYOU MOUNTAINS OF SOUTHWESTERN OREGON Trevor W. Fetz,i Stewart W. Janes, and Heidi Lauchstedt Department of Biology, Southern Oregon University, Ashland, OR 97520 US. A. Abstract, — Great Gray Owls {Strix nebulosa) were located at 92 sites, 1991-97, on the northern slopes of the Siskiyou Mountains of southwestern Oregon. We characterized habitat at the 28 sites including locations with nests (10), recently-fledged young (5), pairs (10), and multiple-year observations (3). Unique to Great Gray Owl populations in western North America, 90% of the nests and 63% of all observations occurred below 915 m in elevation. Owls occurred most often in stands of mature and old growth Douglas-fir {Pseudotsuga menziesii) adjacent to open canopy woodland, chaparral, and natural meadows. Oregon white oak {Quercus garryana) was the dominant tree in the woodlands. Three species associated with woodlands and meadows, California voles {Microtus californicus) , moles { Scapanus spp.) , and Botta’s pocket gophers {Thomomys bottae) comprised 73.3% of the prey items identified. While all nests were near some relatively open habitat, only four of 10 nests were within 800 m of recent clearcuts. All nests were stick platforms in live trees. The distribution of Great Gray Owls in the Siskiyou Mountains suggests that they may have a broader distribution in the western United States than previously reported, particularly at elevations below 915 m. Key Words: Great Gray Owl, Strix nebulosa; food habits; habitat selection; nest sites; Oregon; Siskiyou Moun- tains. GARACTERISTICAS DEL hAbITAT DEL GRAN BUHO GRIS EN LAS MONTANAS SISKIYOU DEL SURESTE DE OREGON Resumen. — Los grandes buhos grises {Strix nebulosa) fueron localizados en 92 sitios entre 1991—97, en las laderas septentrionales de las Montanas de Siskiyou en el sureste de Oregon. Caracterizamos el habitat en los 28 sitios con nidos (10), volantones (5), parejas (10), y observaciones multiples anuales (3). Solo para las poblaciones de Grandes Buhos Grises en el oeste de Norteamerica, el 90% de los nidos y el 63% de todas observaciones ocurrieron por debajo de 915 m de elevacion. Los buhos se presentaron mas frecuentemente en perchas en abetos de Douglas {Pseudotsuga menziesii) maduros y viejos, adyacentes a claros de dosel del bosque, chaparrales, y praderas naturales. El roble bianco de Oregon ( Quercus garryana) fue el arbol dominante en los bosques. Tres especies asociadas con bosques y praderas, Microtus californicus, Scapanus spp. y Thomomys bottae comprendieron el 73.3% de las presas identificadas. Todos los nidos estuvieron cerca de algun habitat relativamente abierto, solo cuatro de 10, estaban dentro de 800 m en claros de bosque recientes. Todos los nidos eran plataformas de palos en arboles vivos. La distribucion de los grandes buhos grises en las Montanas de Siskiyou sugiere que estos pueden tener una distribucion mas amplia en el oeste de los Estados Unidos de lo que previamente se conocia, especialmente en elevaciones por debajo de 915 m. [Traduccion de Cesar Marquez] In the western United States, the Great Gray Owl {Strix nebulosa) is a reclusive inhabitant of boreal forests (Franklin 1988). The species is considered an uncommon resident with an uneven distribu- tion extending from the crest of the Cascade Mountains in Washington and Oregon, east to the 1 Present address: Department of Biology, New Mexico State University, Las Cruces, NM 88003; e-mail address: tfetz@nmsu.edu Rocky Mountains of western Montana and north- western Wyoming (Bull and Duncan 1993, Duncan and Hayward 1994). These owls also occur further south in the Sierra Nevada Mountains of California (Winter 1986). Oregon and California populations are associ- ated with a variety of mature and old-growth co- niferous forest habitats near mountain meadows above 915 m. These include true firs (Ato^spp.), ponderosa and lodgepole pines {Pinus ponderosa 315 316 Fetz et al. VoL. 37, No. 4 Figure 1. Distribution of Great Gray Owl sightings (1991-97) in the Siskiyou Mountains of southwestern Oregon. Solid circles represent nests, solid squares represent recently-fledged young, pairs, and multiple-year sightings, and open circles represent single observations of individuals. Shaded areas represent surveyed locations >915 m, and crosshatched areas represent surveyed locations <915 m. and P. contorta), and Douglas-fir {Pseudotsuga men- ziesii; Winter 1980, 1986, Bryan and Forsman 1987, Bull et al. 1988b, Goggans and Platt 1992). Great Gray Owls use montane meadows as pre- ferred hunting areas (Winter 1986) as well as open stands of mature forests with grass as the dominant ground cover (Bull et al. 1988b) and clearcuts (Goggans and Platt 1992). Denser stands of trees are used less frequently. Nests in the western United States are often in natural depressions in broken-topped snags, but stick nests and artificial platforms are also used (Bull and Henjum 1990, Bull and Duncan 1993). Typically, nests are located within 0.3 km of a meadow or clearcut (Winter 1986, Bryan and Fors- man 1987). In 1991, a population of Great Gray Owls was discovered in the Siskiyou Mountains of southwest- ern Oregon, west of the previously described range of the species (Johnsgard 1988, Bull and Duncan 1993). The owls in this new geographic area are distinctive because they occur below 915 m in el- evation. Therefore, we set out to better understand the habitat used by these owls. In this study, we characterize the habitat and nesting sites occupied by these Great Gray Owls, which occur 34—85 km west of the nearest previously-knovm nesting site. Study Area and Methods In 1991, biologists from the Bureau of Land Manage- ment (BLM) and United States Forest Service (USFS) began to encounter Great Gray Owls in the Siskiyou Mountains, including several occupied nests. In 1996 and 1997, BLM and USFS biologists surveyed all habitat in a 117 600 ha area for this species in the Siskiyous (Fig. 1). All 75 observations in the surveyed area plus 17 addition- al sightings outside the survey area were recorded on 7.5 minute United States Geologic Survey (USGS) topo- graphic maps. Surveys were conducted according to pro- cedures outiined in Huff et al. (1996). This protocol di- rected that surveys be done: (1) within the range of the Spotted Owl {Strix occidentalis) , (2) at elevations above 915 m, (3) within mature stands (80+ years) with >60% canopy cover, and (4) within 305 m of a natural meadow of more than 4 ha. The study area includes a diverse mixture of forest types, ranging from the Interior Valley Zone at the lowest elevations (400 m) to the mountain hemlock {Tsuga mer- tensiana) zone at the highest (2300 m; Franklin and Dyrness 1973). The dominant tree species along the gra- dient beginning at the lowest elevations include Oregon white oak ( Quercus garryana ) , ponderosa pine, Pacific ma- December 2003 Great Gray Owl Habitat 317 drone {Arbutus menziesii), Douglas-fir, California black oak ( Q. kelloggii) , white fir {Abies concolor) , red fir (A. mag- nifica), and mountain hemlock. From the 92 locations recorded by the BLM and USFS between 1991 and 1997, we selected the 28 sites with ei- ther nests, recently-fledged young, pairs, or multiple-year observations for habitat analysis. We defined pairs as two birds in adult plumage seen in the same location at the same time simultaneously between March and August. Because the location of a nest was not identified in the latter three categories, they are referred to as activity cen- ters without nests. “Activity centers” may be in nesting habitat, but they potentially include foraging and roost- ing habitat too. By excluding single observations of lone birds, we minimized the potential of including less im- portant habitats used by transient birds. In addition to the primary study area, we included two sites, one with a nest, located below 915 m in the foothills of the Cascades about 30 km northeast of Medford, Oregon (Fig. 1). We considered these sites because the habitat was similar to that occupied by Great Gray Owls in the Siskiyou Moun- tains. Also, these documented use areas are important because the current federal protocol for the Northwest Forest Plan calls for surveys for Great Gray Owls above 915 m (Huff et al. 1996). At known nest sites, we measured habitat features cen- tered on the nest tree. The actual nest tree at one nest site could not be relocated. For this site, we sampled hab- itat beginning at a point equidistant between the three likely nest trees, all of which were within 10 m of one another. For activity centers, we measured habitat fea- tures centered from the point of observation indicated by government biologists on 7.5 minute USGS topo- graphic maps, coupled with their detailed description of the location of the bird. The sampling area consisted of a 19.95-m radius circle (0.125 ha) plus four additional 19.95-m radius plots centered 100 m from the center of the central plot in each of the cardinal directions. We recorded the identity and diameter at breast height (DBH) of live trees and snags >8 cm DBH within each plot. We measured slope and aspect at the center of each plot using a clinometer and a compass. We calculated the foliage profile at the center of each plot by measuring vegetation density in each of the four cardinal directions according to MacArthur and MacArthur (1961). We vi- sually estimated percent canopy closure. We measured the height of each nest and the nest tree with a clinometer. In addition, tree species, DBH, nest type, and nest placement relative to the tree were re- corded. We also counted the number of stick nests within 50 m of the nest tree to determine their availability and to help identify the species that may have built the nest used by the owls. For example, a large number of stick nests within a 50 m radius would tend to eliminate accip- iters as the nest builders (Reynolds and Wight 1978). Small mammal species are often restricted to particular habitats (Verts and Carraway 1998). We analyzed the con- tents of Great Gray Owl pellets to gain further clues to owl habitat use. We estimated the number of prey from counts of skulls and mandibles. We used BLM and USFS aerial photos to assess habitat within 0.8 km (2.0 km^) of each nest or observation. For each site, we used aerial photos from the year that most closely corresponded to the year of observation (1991- 96) . We verified with government biologists that no hab- itat alteration occurred in the intervening years. After determining the scale of each photo by comparison with USGS topographic maps, we placed a 67 X 67 m grid (113 points) over the nest or point of observation and assigned each grid point to one of the following vegeta- tion categories: closed-canopy coniferous forest, open- canopy forest, chaparral, or open areas. We defined open-canopy forest as Oregon white oak and oak/ madro- ne woodland and thinned stands (hereafter partial cuts) <10 yr old. Chaparral consisted of habitats dominated by wedgeleaf ceanothus ( Ceanothus cuneatus) , whiteleaf man- zanita {Arctostaphylos vicida) and poison oak {Toxicoden- dron diversiloba) . Open areas included natural meadows, clearcuts <10 yr old, agricultural fields, and roads. We also measured the distance to the nearest clearcut or par- tial cut <10 yr of age and meadow >2 ha in extent. We did not consider clearcuts >10 yr of age to be open areas because the growth of young conifers and shrubs ap- proached a closed canopy condition on most sites by this time. Level of significance for all tests was set at a = 0.05 Comparison of size composition of trees at nest sites and activity centers was initially evaluated with a Multivariate Analysis of Variance (MANOVA) using Statistica for Win- dows (1993). However, the assumption of equal covari- ance matrices was violated, thus we compared the size classes independently. Due to multiple comparisons, we used a sequential Bonferroni correction (Rice 1989). When comparing the 30 habitat features between nests and activity centers (Table 2), we employed the Mann- Whitney U-test when the data were not normally distrib- uted, again using a sequential Bonferroni correction. We evaluated the elevational distribution of Great Gray Owls relative to 915 m and habitat data involving percentages using a Gtest. Results Great Gray Owls were detected at 92 locations, including 10 sites with nests, five sites with fledg- lings but no identified nests, 10 sites with pairs but no breeding confirmed, and 67 sites with single owls confirmed in one or more years (Fig. 1). Of the 92 sites, 63% were below 915 m (Fig. 2). Of the 25 sites where nests, fledglings, or pairs were confirmed, 80% were below 915 m, with a mean elevation of 904 ± 397 (±SD). Within the area sur- veyed by government biologists, a similar pattern existed indicating greater abundance below 915 m (Gi - 11.32, P < 0.001 and Gj = 9.25, P < 0.05 for all sites and those with a nest, fledglings, and pairs; respectively) . All sites were located on public lands administered by the Medford District of the Bureau of Land Management, or the Rogue River National Forest. Ten nests were found during the study, and an- other was found in 1998, when conducting follow- 318 Fetz et al. VoL. 37, No. 4 Elevation (m) Figure 2. Elevational distribution of Great Gray Owls in the Siskiyou Mountains. “Single” represents locations of individuals observed once. “Pair” indicates locations with pairs and/or multiple-year sightings. “Nest” represents observed nests or locations of recently-fledged young. up searches of previously used nesting sites. None of these nests were located closer than 4 km of each other. Of these, 10 were in Douglas-fir and one was in a ponderosa pine. Of 10 cases where the specific nest was identified, five were stick nests situated on lateral limbs against tree trunks, and five were stick nests in the forked top of trees. None of the nests were reused the following year. Nests were located in one of the largest trees at each site (Tables 1 and 2). The number of stick nests within 50 m of nests used by owls ranged from 2-14 {x = 5.9). Incubating or brooding Great Gray Owls were observed between 25 March and 12 May; the earliest fledglings were observed 14 May. Great Gray Owls most frequently nested in late- successional stands dominated by Douglas-fir locat- ed near natural forest edges (Table 2). Nest plots typically included large trees at low densities with numerous smaller understory trees. Of the five subplots sampled at each site, basal area averaged highest on the nest plot and dropped to less than 50% on the lowest-ranked peripheral plot (Fig. 3). The decrease in Douglas-fir basal area and the cor- responding increase in Oregon white oak basal area between the nest plot and the lowest-ranked peripheral plot also indicated a change in habitat within 120 m of the nest tree. Oregon white oak tends to occur in single-species or hardwood-dom- inated open-canopy stands and is uncommon with- in areas dominated by Douglas-fir. Forest composition at nest sites was similar in most respects to activity centers without nests. The Table 1. Characteristics of 10 nest trees used by Great Gray Owls in the Siskiyou Mountains of southwestern Oregon, 1991-97. Feature Mean SD Range DBH (cm) of nest tree 64.0 24.0 34-106.5 Nest height (m) 20.9 5.1 16-31 Tree height (m) Number of stick nests within 35.3 6.4 2-45 50 m 5.9 3.6 2-14 distribution of trees in regard to six size classes did not differ (P > 0.767). Of the 30 variables exam- ined (Table 2) , no significant differences between nest sites and activity centers were detected, and only two were indicated when not using the se- quential Bonferroni correction (Oregon white oak density and the size of conifer snags). Because of the similarities in habitat features between nest sites and activity centers, we feel justified combin- ing the two to characterize the habitat of Great Gray Owls in the Siskiyou Mountains (Table 2). Both nest and all sites combined contained about 45% closed canopy forest within a 0.8 km radius (Table 3). Only four of ten nest sites and 15 of all 28 sites had clearcuts within 0.8 km. Clearcuts never accounted for more than 10% of the area at the ten nest sites. Aspect was predominantly north, with 80% of nest sites and 67% of activity centers facing north. Overall, 20 of 28 sites had a northerly aspect {P < 0.02, binomial probability of a result this extreme or greater). Pellets and pellet fragments were collected at 11 locations, including seven nest sites and four roosts. Of 165 prey items identified, 79.4% were located beneath nests. California voles {Microtus ca- Ufornicus) and moles {Scapanus were the most common prey items, comprising 63.0% of the in- dividuals identified (Table 4). Botta’s pocket go- pher {Thomomys bottae) was the only other species accounting for >5% of the prey items. Discussion Great Gray Owls of the Siskiyou Mountains most often occur in mature Douglas-fir forest close to Oregon white oak or oak/madrone woodland, chaparral, or natural meadows. The use of mature and old growth forests adjacent to open areas is similar to findings of other studies in the western United States (Forsman and Bryan 1984, Winter December 2003 Great Gray Owl Habitat 319 Table 2. Habitat characteristics in the vicinity of 10 nest sites and 18 additional sites with fledged young, pairs, or multiple-year occupancy in the Siskiyou Mountains, 1991-97. Nest Sites Ait, Sites^ Feature Mean SD Range Mean SD Range Tree density (No. /ha) Douglas-6r 251 156 75-630 274 165 0-630 White fir 20 64 0-64 25 68 0-245 Ponderosa pine 11 12 0-32 15 23 0-106 Other conifer 10 19 0-56 27 63 0-250 Pacific madrone 74 75 0-203 61 63 0-203 Oregon white oak 105 162 0-547 49 106 0-547 California black oak 32 30 0-77 29 33 0-142 Other hardwood 13 16 0-50 16 41 0-216 Total 518 254 262-1062 497 211 205-1062 Snag density (No. /ha) Conifer 41 16 22-74 43 29 0-138 Hardwood 54 47 18-171 36 36 0-171 Total 95 55 40-219 78 52 0-219 DBH live trees (cm) Douglas-fir 31 10 18-47 30 10 13-53 White fir 28 0 28 21 8 10-28 Ponderosa pine 36 7 24-46 36 23 16-125 Other conifer 28 2 27-30 23 13 9-55 Pacific madrone 21 8 4-33 20 6 4-33 Oregon white oak 16 5 11-26 16 4 11-26 California black oak 20 6 14-31 19 5 12-31 Other hardwood 17 6 12-27 15 5 8-27 Total 25 5 15-32 25 5 15-32 DBH snags (cm) Conifer 26 7 13-37 23 7 11-37 Hardwood 15 2 13-19 15 3 10-25 Total 20 3 13-27 19 4 11-27 Distance to water (m) 405 (10)2 163 12-596 372 (28) 252 12-131 Distance to timber sale (m) 129 (4) 78 40-184 145 (18) 67 40-251 Distance to opening (m) 99 (10) 89 22-313 103 (28) 87 0-325 % canopy closure 65 14 42-81 61 19 9-87 % slope 37 16 8-59 36 15 8-65 Elevation (m) 704 161 512-1006 901 391 427-2262 1 “All sites” includes nest sites and other activity centers (locations with recently-fledged young, pairs, and multiple-year observations where a nest has not been located) . ^ Numbers in parentheses represent the number of sites at which a feature occurred within 1 km and were used in calculating the mean. 1986, Franklin 1988, Bull and Henjum 1990, Gog- gans and Platt 1992), but not with respect to ele- vation (<915 m) and not near Oregon white oak and chaparral habitat types. This suggests a wider elevational and habitat distribution than currently understood. The nature of the openings used by Great Gray Owls for foraging varies throughout its range in the western United States. In California, owls for- aged primarily in natural meadows (Winter 1986). In contrast, owls in the Rocky Mountains and the central Cascades of Oregon foraged primarily in clearcuts (Franklin 1988, Goggans and Platt 1992). Only in northeastern Oregon, did birds hunt reg- 320 Fetz et al. VoL. 37, No. 4 Nest Per. 1 Per. 2 Per. 3 Per. 4 Figure 3. Basal area (m^/ha) of Douglas-fir, Oregon white oak, and all trees combined in the nest plot and peripheral plots (1-4) at 10 Great Gray Owl nest sites in the Siskiyou Mountains. Peripheral plots are ranked from greatest to least total basal area. ularly in open canopy forests (Bull et al. 1988a, 1988b, 1989). In the Siskiyous, the openings in the vicinity of owl observations most often included open Oregon white oak and oak/ madrone wood- lands as well as chaparral and grassy meadows. Clearcuts were uncommon near owl locations, and based on prey items identified from pellets, do not appear important as foraging areas. At lower ele- vations in the Siskiyou Mountains, clearcuts tend Table 3. Percent coverage of different vegetation types within a 0.8-km radius (2 km^) centered on the nest or point of observation of Great Gray Owls in the Siskiyou Mountains, 1991-97. Habitat Nest Sites All SlTES^ Closed-canopy coniferous forest 45.5 45.9 Open-canopy forest Oak and oak/madrone woodland 23.8 17.0 Partial/selective cuts 3.8 7.0 Other 0 1.2 Total 27.6 25.2 Chaparral 12.6 10.3 Open Natural meadows 3.7 7.4 Recent clearcuts 3.8 6.1 Agricultural fields 4.0 2.0 Roads 2.2 2.5 Total 13.6 18.0 Other (buildings, yards, farm ponds) 0.7 0.6 ^ “All sites” includes nest sites and other activity centers (loca- tions with recendy-fledged young, pairs, and multiple-year obser- vations where a nest has not been located) . Table 4. Prey items identified in the pellets of 11 pairs of Great Gray Owls inhabiting the Siskiyou Mountains, 1991- 97. Species Individuals Percent of Total Items Percent Occurrence^ Mole Scapanus sp. 49 29.7 82 Shrew-mole Neurotrichus gibbsii 3 1.8 27 Shrew Sorex sp. 3 1.8 18 Northern flying squirrel Glaucomys sabnnus 1 0.6 9 Botta’s pocket gopher Thomomys bottae 16 9.7 55 Mazama pocket gopher Thomontys mazama 1 0.6 9 Deer mouse Peromyscus maniculatus 1 0.6 9 Mouse Peromyscus sp. 2 1.2 9 Dusky-footed wood rat Neotoma fuscipes 1 0.6 9 Wood rat Neotonia sp. 1 0.6 9 Creeping vole Microtus oregoni 3 1.8 27 California vole Microtus californicus 56 33.9 82 Townsend’s vole Microtus townsendii 5 3.0 18 Vole Microtus sp. 13 7.9 45 Red-backed vole Clethrionomys occidentalis 2 1.2 9 Norway rat Rattus norvegicus 1 0.6 9 Unidentified mammal 6 3.6 18 Dark-eyed junco Junco hyemalis 1 0.6 9 Total 165 99.8 ' Percent occurrence represents the percentage of sites at which a given prey type was encountered. December 2003 Great Gray Owl Habitat 321 to be dominated by brush and young conifers be- fore a dense herbaceous layer develops. The Great Gray Owl’s principal prey, California voles, moles, and Botta’s pocket gophers prefer grasslands and other habitats with abundant herbaceous vegeta- tion such as that found in Oregon white oak and oak/ madrone woodlands, as well as in grassy mead- ows (Bailey 1936, Ingles 1965, Maser and Storm 1970). The diet of Great Gray Owls on our study area was similar to many previous studies, in that voles, gophers, and other small terrestrial mammals were the dominant source of food (Mikkola and Sulkava 1970, Mikkola 1972, Winter 1986, Franklin 1988, Bull and Henjum 1990) . The abundance of moles in the diet, however, is higher than reported in other studies. The dominance of these small ter- restrial mammals in the diet suggests that the owls were foraging in the more open oak/madrone woodland and grassy areas. Because of their large size (410-465 mm wing chord; Johnsgard 1988), we believe that Great Gray Owls are unable to for- age efficiently in dense forest habitats, and are best adapted to hunt edges and open forests. The elevational distribution of Great Gray Owls in the Siskiyou Mountains is unique in the western United States. In other locations, almost all owls have been reported above 915 m (Forsman and Bryan 1984, Franklin 1988, Bull and Henjum 1990). The few observations below 915 m were at- tributed to dispersing juveniles, non-breeders, and wintering birds (Winter 1986, Goggans and Platt 1992). The lowest reported nesting was 912 m (Goggans and Platt 1992). By comparison, only one of the 1 1 nests in this study was above 825 m. The distribution of suitable habitat in the Siski- yous may explain the large number of owls ob- served at low elevations. Below 915 m, mature Douglas-fir forests on north-facing slopes often oc- cur in close proximity to Oregon white oak-domi- nated woodlands, chaparral and/ or natural mead- ows on adjacent south-facing slopes. In these areas, Douglas-fir stands offer roosting and nesting sites while the open woodlands and meadows offer op- portunities for foraging. Oak woodlands tend to support a dense herbaceous layer, and numerous hunting perches which are important to Great Gray Owls (Bull and Henjum 1990). Above 915 m in the Siskiyous, vegetation is dominated by conif- erous forests, with few oak woodlands or open grasslands, though clearcuts are widespread. Great Gray Owls appear to be opportunistic re- garding nest selection, using whichever nest type is available, including artificial nest platforms (Nero 1980, Mikkola 1983, Bull and Duncan 1993, Whit- field and Gaffney 1997, Duncan 1997). At low el- evations in the Siskiyous, platform nests were abun- dant in most stands where owls nested. Owls nested in one of the largest trees (top 10%) in each stand illustrating the importance of mature stands. The location and structure of stick nests used by the owls led us to believe that most of the nests were constructed by western gray squirrels {Sciurus gn- seus ) , which tend to be associated with oak wood- lands or with the interface between oak woodlands and coniferous forests (Bailey 1936, Gross 1969, Maser 1998, Verts and Carraway 1998). This habitat is common below 1000 m on the northern slopes of the Siskiyou Mountains. The abundance of po- tential nest sites may explain why nests were not reused in subsequent years. Another potential factor affecting the elevatio- nal distribution of Great Gray Owls is temperature. Winter (1986) speculated that Great Gray Owls do not nest at lower elevations in the southern por- tion of their range because they are adapted to boreal forests and are physiologically unable to oc- cupy areas with high temperatures. Our observa- tions suggest that the southern limits of the range may have more to do with the distribution of suit- able habitat than with temperature per se. All nests that we located occurred at elevations where mean maximum temperatures exceeded 32°C during the summer months, and where temperatures on some summer days surpassed 40°C (Taylor 2001). Cloud cover and precipitation provide limited relief dur- ing the breeding season with less than 3 cm of pre- cipitation in each April and May and even less in the summer months (Taylor 2001). However, all nests in the Siskiyou Mountains were shielded from direct sunlight by canopy cover, in contrast to the exposed broken-top snag nests found in California (Winter 1986). Great Gray Owls regularly occur below 915 m in the Siskiyou Mountains where suitable habitat is available. As such, a reassessment of the current protocol used by the USFS and BLM to monitor Great Gray Owls in the Pacific Northwest in terms of conducting surveys only above 915 m (3000 ft) seems appropriate. Thus, Great Gray Owls may oc- cur elsewhere in the western United States at ele- vations lower than previously recorded where shad- ed nesting sites are available. 322 Fetz et al. VoL. 37, No. 4 Acknowledgments We thank Steve Godwin, Laura Finley, Dave Clayton, and Linda Hale for providing locations of Great Gray Owl sightings. We also thank Steve Godwin for generat- ing the map of the study area with owl sightings. Michelle Rowe provided valuable assistance in the field. We thank Evelyn Bull, Eric Forsman, and the referees for their helpful comments in the preparation of this manuscript. Literature Cited Bailey, V. 1936. The mammals and life zones of Oregon. USD A Bureau of Biological Survey, Washington, DC U.S.A. Bryan, T. and E.D. Forsman. 1987. Distribution, abun- dance, and habitat of Great Gray Owls in south cen- tral Oregon. Murrelet 68:45—49. Bull, E.L. and J.R. Duncan. 1993. Great Gray Owl {Strix nebulosa). Pages 1-16 in A. Poole and F.B. Gill [Eds.], The birds of North America. Vol. 2. Academy of Nat- ural Sciences, Philadelphia, PA and American Orni- thologists’ Union, Washington, DC U.S.A. and M.G. Henjum. 1990. Ecology of the Great Gray Owl. Gen. Tech. Rep. PNW-GTR-265. USDA For- est Service, Pacific Northwest Research Station. Port- land, OR U.S.A. , , AND R.S. Rohweder. 1988a. Home range and dispersal of Great Gray Owls in northeastern Oregon./. Raptor Res. 22:101-106. , , AND . 1988b. Nesting and foraging habitat of Great Gray Owls./. Raptor Res. 22:107-115, , , AND . 1989. Diet and optimal for- aging of Great Gray Owls. /. Wildl. Manage. 53:47-50. Cross, S.P. 1969. Behavioral aspects of western gray squir- rel ecology. Ph.D. dissertation, University of Arizona, Tucson, AZ U.S.A. Duncan, J.R. 1997. Great Gray Owls {Strix nebulosa nebu- losa) and forest management in North America: a re- view and recommendations./. Raptor Res. 31:160-166. AND P.A. Hayward. 1994. Review of technical knowledge: Great Gray Owls. Pages 159-175 in G.D. Hayward and J. Verner [Eds.], Flammulated, Boreal, and Great Gray Owls in the United States: a technical conservation assessment. USDA Forest Service Gen. Tech. Rep. RM-253, Fort Collins, CO U.S.A. Forsman, E.D. and T. Bryan. 1984. Distribution, abun- dance, and habitat of Great Gray Owls in south-cen- tral Oregon, Ore. Dept, of Fish and Wildlife. Non- game Wildlife Project #84-3-05, Corvallis, OR U.S.A. Franklin, A.B. 1988. Breeding biology of the Great Gray Owl in southeastern Idaho and northwestern Wyo- ming. Conc?<>r 90:689-696. Franklin, J.R and C.T. Dyrness. 1973. Natural vegetation of Oregon and Washington. USDA Forest Service Gen. Tech. Rep. PNW-8. Portland, OR U.S.A. Goggans, R. and M. Platt. 1992. Breeding season ob- servations of Great Gray Owls on the Willamette Na- tional Forest, Oregon. Oreg. Birds 18:35-41. Huff, M., J. Henshaw, and E. Laws. 1996. Great Gray Owl survey status and evaluation of guidelines for the northwest forest plan. USDA Forest Service and USDl Bureau of Land Management. Unpubl. report. Port- land, OR U.S.A. Ingles, L.G. 1965. Mammals of the Pacific States, Stan- ford Univ. Press, Stanford, CA U.S.A. Johnsgard, P.A. 1988. North American owls. Smithson. Inst. Press, Washington, DC U.S.A. MacArthur, R.H. and J.W. MacArthur. 1961. On bird species diversity. Ecology 42:594—598. Maser, C. 1998. Mammals of the Pacific Northwest. Oregon State Univ. Press, Corvallis, OR U.S.A. and R.M. Storm. 1970. A key to Microtinae of the Pacific Northwest (Oregon, Washington, Idaho). Oregon State Univ. Book Stores, Corvallis, OR U.S.A. Mikkola, H. 1972. Food of Great Grey Owls in the Lap- land Reserve, USSR. Brit. Birds 65:31-32. . 1983. Owls of Europe. Buteo Books, Vermillion, SD U.S.A. and S. Sulkava. 1970. Food of Great Grey Owls in Fenno-Scandia. Brit. Birds 62:23-27. Nero, R.W. 1980. The Great Gray Owl: phantom of the northern forest. Smithson. Inst. Press, Washington, DC U.S.A. Reynolds, R.T. and H.M. Wight. 1978. Distribution, den- sity, and productivity of accipiter hawks breeding in Oregon. Wilson Bull. 90:182—196. Rice, W.R. 1989. Analyzing tables of statistical tests. Evo- lution 43:223-225. Statistica for Windows. 1993. Release 4.5. StatSoft, Inc. Tulsa, OK U.S.A. Taylor, G.H. 2001, Oregon Weather Summary. Oregon Climate Service. Oregon State University, Corvallis, OR U.S.A. Verts, B.J. and L.N. Carraway. 1998. Land mammals of Oregon. Univ. California Press, Berkeley, CA U.S.A. Whitfield, M.B. and M. Gaffney. 1997. Great Gray Owl {Strix nebulosa) breeding habitat use within altered forest landscapes. Pages 601-606 mJ.R. Duncan, D.H. Johnson, and T.H. Nicholls [Eds.], Biology and con- servation of owls of the northern hemisphere: sym- posium proceedings. Gen. Tech. Rep. NC-190, Win- nipeg, MB Canada. Winter, J. 1980. Status and distribution of the Great Gray Owl in California. California Dept, of Fish and Game, Federal Aid Restoration. Final Report W-54-R-12, Sac- ramento, CA U.S.A. . 1986. Status, distribution and ecology of the Great Gray Owl {Strix nebulosa) in California. M.S. the- sis, San Francisco State University, San Francisco, CA U.S.A. Received 13 November 2001; accepted 12 May 2003 Former Associate Editor: Cole Crocker-Bedford J Raptor Res. 37(4):323-329 © 2003 The Raptor Research Foundation, Inc. NEST-SITE CHARACTERISTICS AND LINEAR ABUNDANCE OF CLIFF-NESTING AMERICAN KESTRELS ON SAN CLEMENTE ISLAND, CALIFORNIA Brian L. Sullivan/ Eric L. Kershner/ Sean R Finn,^ Anne M. Condon, Douglass M. Cooper, and David K. Garcelon^ Institute for Wildlife Studies, 2515 Camino del Rio South, Ste. 334, San Diego, CA 92108 U.S.A. Abstract. — ^American Kestrels {Falco sparverius) are typically secondary-cavity nesters, and use of natural cliff cavities for nest sites is less-commonly reported. On San Clemente Island (SCI) , California, however, American Kestrels nest primarily on cliffs in major canyons (93%), to a lesser extent on sea cliffs (4%), as well as in man-made structures (3%). We located and mapped 99 American Kestrel territories on SCI, and recorded II nest-site characteristics at 40 cliff nests during 2001-02. Nest cliffs were typically fractured igneous rock with mean height of 16.1 m ± 1.8 SE. Mean slope of nest cliffs was vertical (x = 91°). Nest cliffs and cavities were significantly oriented to the southeast, away from the prevailing wind direction (NW) . In eight canyons, where we believe that we found all occupied American Kestrel territories, the mean linear abundance was 2.1 pairs/km, greater than most published estimates. Con- trary to most previous studies, no American Kestrels nested in tree cavities despite their presence in SCI canyons. The absence of cavity-excavating breeding birds from the island likely restricts kestrels to nesting in naturally-formed cavities and man-made structures. Key Words: American Kestrel; Falco sparverius; California; cliff-nesting, nest abundance, nest site characteris- tics; San Clemente Island. CARACTERISTICAS DE LOS SITIOS NIDO YABUNDANCIA LINEAR DE CERNICALOS QUE ANI- DAN EN ACANTILADOS DE LA ISLA SAN CLEMENTE, CALIFORNIA Resumen. — Los cermcalos {Falco sparverius) anidan tipicamente en cavidades usadas, el uso de cavidades naturales en los acantilados como sitios nido se reporta menos comunmente. En la Isla de San Clemente (ISC) , California, sin embargo, los cernicalos anidan primordialmente en los precipicios de los grandes canones (93%), y en menor grado en los acantilados marines (4%), al igual que en estructuras cons- truidas por el hombre (3%). Localizamos y colocamos en un mapa 99 territorios de cernicalos en la ISC, y registramos 11 caracteristicas de los sitios nido en 40 nidos de acantilados durante 2001-02. los nidos de los precipicios fueron tipicamente roca ignea fracturada con una altura media de 16.1 m ± 1.8 SE. La pendiente promedio de los nidos en acantilados fue vertical {x = 91°). Los nidos y las cavidades estuvieron significativamente orientadas hacia el suroriente, lejos de la direccion prevaleciente del viento (NW). En ocho canones, donde creemos que encontramos todos los territorios de los cer- nicalos ocupados, el promedio de la abundancia linear fue 2.1 parejas/km, mucho mayor que la mayoria de las estimaciones antes reportadas. Contrario a la mayoria de los estudios previos, ningun cernicalo anido en las cavidades de los arboles a pesar de su presencia en los canones de la ISC. La ausencia en la isla de aves que aniden en cavidades excavadas por ellos mismos, probablemente restringe a los cernicalos a anidar en cavidades formadas naturalmente o en estructuras construidas por el hombre. [Traduccion de Cesar Marquez] ^ Present address: PRBO Conservation Science, 4990 Shoreline Hwy., Stinson Beach, CA 94970 U.S.A. ^ Corresponding author’s e-mail address: kershner@iws.org Present address: USGS, Forest and Rangeland Ecosystem Science Center, 970 Lusk St., Boise, ID 83706 U.S.A. Present address: Institute for Wildlife Studies, P.O. Box 1104, Areata, CA 95518 U.S.A. 323 324 Sullivan et al. VoL. 37, No. 4 American Kestrels {Falco sparvetius) are the most widely distributed and abundant North American falcon. The species is a secondary-cavity nester typ- ically using woodpecker-excavated or naturally- formed cavities in trees, cliffs, and man-made struc- tures (Smallwood and Bird 2002). Although numerous studies described nest box populations (e.g., Smallwood and Collopy 1991, Rohrbaugh and Yahner 1997, Smallwood and Wargo 1997), and extensive work was conducted on tree-nesting populations (Smith et al. 1972, Balgooyen 1976), few quantitative data are available regarding this species’ nesting habits in natural cliff cavities (but see Smith et al. 1972, De Lucca and Saggesse 1993). We measured 11 nest site characteristics for 40 confirmed American Kestrel nest cavities in nat- ural cliffs on San Clemente Island (SCI), Califor- nia. Although several studies have reported density for tree nesting and nest box populations (Craig and Trost 1979, Smallwood and Bird 2002), esti- mates of density and abundance for cliff-nesting American Kestrels in North America have not been previously reported. Therefore, we present mean linear abundance of breeding pairs in eight can- yons on SCI. Methods Study Area. San Clemente Island is the southernmost of the California Channel Islands (32°50'N, 118°30'W), located in Los Angeles County ca. 92 km west of Palos Verdes, CA. SCI is administered by the U.S. Navy and is used for active military training as part of the Southern California Offshore Range; however, the U.S. Navy has an extensive environmental program on the island for the protection and conservation of natural resources (USDN, SWDIV 2001). The island is 44 km long and 2.4- 6.4 km wide, encompassing ca. 14 764 ha. Elevation rang- es from sea level to 599 m. The island consists of a central plateau incised by canyons along the western and eastern slopes. The east side is a precipitous escarpment, de- scending from the plateau to the Pacific Ocean; the west side consists of a series of more gentle marine terraces formed as a result of periodic uplifting and erosion from wave action (Vedder and Howell 1980). Canyon bottoms typically contain little water, with continuous flow restrict- ed to rain events. Annual temperature ranges from 10— 22°C and mean annual precipitation is 15.7 cm (Olms- tead 1958, Vedder and Howell 1980). Current vegetation on SCI includes mixed native and non-native grasslands (including Avena, Bromus, and Stipa spp.) on the upper terraces; maritime desert scrub con- taining California sagebrush {Artemesia californica) , morn- ing glory ( Calystegia macrostegia) , prickly pear ( Opuntia lit- toralis), cholla (O. prolifera), snake cholla {Bergerocactus emoryii), and wild oats {Avena sp.); and remnant patchy shrubland/woodland characterized by island cherry (Prunus lyonii) , lemonade berry {Rhus integrifolia) , island oak {Quercus tomentella), canyon live oak {Quercus chryso- lepis), and Catalina ironwood {Lyonothamnus floribundus) (Raven 1963, Sward and Cohen 1980, Kellogg and Kel- logg 1994). Nest-site Characteristics. We located and mapped 99 American Kestrel nest sites on SCI during 2001-02 (Fig 1). We surveyed the island on foot or from a boat during the breeding season (February-June). Territories were considered occupied when a courting adult pair (e.g., prey exchanges, copulation) was present. Nest sites were located by observing adult behavior. When possible, we visually located nests by following an adult flying to its cavity. Nest cavities were defined as occupied when a prey delivery was observed. Nest sites were plotted on topo- graphic maps generated from ArcView 3.2a (Environ- mental Systems Research Institute, Redlands, CA). If a nesting territory was occupied in two consecutive years, only the first nest detected was included in the nest-site characteristics analysis. Nest sites were classified as in a canyon cliff, sea cliff, tree, or artificial structure. Canyon cliffs were found with- in interior canyons. Sea cliffs bordered the ocean. Arti- ficial structures included buildings, junk piles, and aban- doned trailers. We measured nest-site characteristics for all occupied nests where the exact nest cavity was directly visible {N = 40). Standard error of the mean is reported. We mea- sured wall height, cliff height, and distance to the canyon bottom using a laser rangefinder and clinometer from the closest accessible perpendicular vantage. Wall height and cliff height were measured on a vertical axis through the nest cavity. Wall height was defined as the area from the canyon bottom to the canyon rim or terrace edge, including vegetated slopes, sheer cliff faces, and rock out- croppings. Cliff height was defined as the facet of the wall (e.g., rock outcropping, sheer face) containing the nest cavity, from its bottom to its top. In some cases when walls were near vertical, wall height and cliff height were identical. Distance to the canyon bottom was measured from the nest to the canyon bottom. We quantified general characteristics of nesting cliffs by estimating percent slope, cliff complexity, and cliff composition. Two to three observers visually estimated percent slope and then we averaged all estimates. In the field, nest cliffs were estimated as a percentage of vertical, where 100% = vertical. Percent slope was then trans- formed into degrees (100% = 90°) for analysis. The com- plexity of the nest cliffs surface was indexed visually and categorized as either flat or convoluted based on the presence of caves, potholes, or jutting rocks. Nest cliff compo.sition was categorized as either solid (no holes or cracks), loose rock (loose granular, scree), fractured (cracked or segmented) or stratified (layered). The amount of vegetation obscuring the entrance of each cav- ity was categorized as clear of vegetation, partially ob- scured by vegetation, or completely obscured by vegeta- tion. Nest cliff aspect and nest orientation (for visible cavi- ties) were measured using the mean compass reading of two or three independent observers. We used Program ORIANA (Kovach 1994) to calculate circular statistics. Standard error of the mean vector is reported. For cliff December 2003 Cliff-nesting American Kestrels 325 Figure 1. Map of 99 occupied American Kestrel nesting territories on San Clemente Island, CA, 2001-02. 326 Sullivan et al. VoL. 37, No. 4 0 ° Cavity 90 " Figure 2. Mean vector analysis of American Kestrel nest cavities, San Clemente Island, CA 2001-02 (Mean vector (|ul) = 147° ± 13° SE, P< 0.001, N= 36). Figure 3. Mean vector analysis of American Kestrel nest cliffs, San Clemente Island, CA 2001-02 (Mean vector ((x) = 155° ± 11° SE, P < 0.001, A = 40). aspect and nest orientation, Rayleigh’s test of uniformity was used for mean vector analysis. Linear Abundance. In each of eight canyons that were thoroughly searched, we determined the number of nest- ing pairs/linear km using ArcView 3.2a. The length of each canyon was calculated using the distance tool in ArcView. Canyon length was defined as the total length ol each canyon from its inception on the upper terrace to its mouth along the ocean. We used a linear measure due to confounding factors involved in determining ex- act area of canyons. The overall mean number of pairs/ linear km was then calculated for the eight canyons. Resui.ts American Kestrels nested primarily in canyons on SCI (Fig. 1). Of 99 territory locations, 92 were on cliffs within interior canyons, whereas four were centered on sea cliffs and three were in man-made structures. Detailed characteristics of nest sites are presented for 40 nests unless otherwise indicated. Mean wall height was 79.4 m ± 7.9 SE, but varied substantially at measured nests. Mean cliff height was 16.1 m ± 1.8 SE. Mean distance from the nest cavity to the canyon bottom was 45.8 m ± 5.8 SE. American Kestrels nested primarily on vertical cliffs (x slope = 91°). Only one of 38 (3%) nesting cliffs was classified as being flat whereas the re- maining 37 (97%) were characterized as convolut- ed. Thirty-six nest cliffs (90%) were categorized as fractured and two (5%) were solid; loose rock and stratified cliffs accounted for 2.5% each. Nest-cavity entrances ranged from being completely clear of vegetation (25; 62%) to completely obscured (4; 10%), with 11 (28%) at least partially obscured by vegetation. Vector analysis of 36 nest cavities (x vec- tor (fx) = 147° ± 13° SE, P < 0.001) and 40 nest cliffs (x vector (p) = 155° ± 11° SE, P < 0.001) indicated that both were oriented to the southeast (Figs. 2 and 3). Mean number of nesting pairs in eight canyons was 2.1 ± 0.2 SE pairs/linear km (range = 1.1-3. 2 pairs/km) (Table 1). Discussion Nest-site Characteristics. American Kestrels are the most widespread breeding raptor on SCI (In- Table 1. Linear abundance of nesting American Kes- trels on San Clemente Island, 2001-02. Canyon Name Mean Length Number of IN KM Pairs/yr Pairs/km Cave Canyon 5.5 11 2.0 Horse Canyon 4.5 8 1.8 Box Canyon 3 5 1.7 Norton Canyon 3.1 8 2.6 Middle Ranch Canyon 2.1 5.5 2.6 Wallrock Canyon 1.7 5.5 3.2 China Canyon 6.3 7 1.1 Horse Beach Canyon 4 6 1.5 Totals 30.2 56 2.1 December 2003 Cliff-nesting American Kestrels 327 stitute For Wildlife Studies unpubl. data) , and are found predominately in canyons with vertical cliffs. The largely cliff-nesting habits of American Kes- trels on SCI have not been commonly reported in previously-described populations. We found Amer- ican Kestrels nesting in interior canyon cliffs de- spite the fact that trees, sea cliffs, and man-made structures were available. Although data on avail- ability of these three cavity types were not collect- ed, our results suggest high usage of canyon cliffs for nest sites. In central Utah, Smith et al. (1972) found only two of 41 nests (5%) in cliffs (aban- doned rock quarries) , the remainder being located in trees and man-made structures. While the ge- ography of Smith et al.’s study site was clearly dif- ferent from that of SCI, similar nesting resources were available for both populations. In addition, cliff nests reported in Smith et al. (1972) were lo- cated in a human-altered landscape, further con- founding potential comparison with those found in naturally-formed cavities in cliffs on SCI. The high incidence of cliff use in our study indicates that American Kestrels will accept natural cliff cav- ities where tree cavities are not available. Several species of woodpeckers visit SCI during the non- breeding season (Jorgensen and Ferguson 1984); however, they do not construct nest cavities. Al- though natural-tree cavities occur on SCI, they are likely uncommon, and some features (e.g., loca- tion, cavity dimensions) may render them unsuit- able for nesting American Kestrels. American Kestrels did not use sea cliffs for nest- ing, despite their abundance on SCI. We hypoth- esize that sea cliff cavities may have undesirable moisture levels due to crashing surf and salt spray, and may be too exposed to wind. Also sea cliff nests may negatively influence fledging success due to their location over water. Nest sites were found on a variety of wall heights, however, nest cavities were confined to sheer rock faces or rock outcroppings within these walls. The lack of American Kestrel nest cavities in anything other than fractured, convoluted rock may be an artifact of the geomorphic composition of SCI. Nest cliffs appear to be highly variable, as is evident in the large range of cliff sizes, wall heights, and elevations above the canyon bottom in our sample. Vertical cliffs seem most likely to be used for nest- ing on SCI and may be associated with protection from mammalian predation (e.g., black rat {Rattus rattus]^ house mouse [Mus musculus], deer mouse [Peromyscus maniculatus] , feral cat [Felis sylvetris] , is- land fox [ Urocyon littoralis ] ) . In contrast to other studies (Rohrbaugh and Yahner 1997, Smallwood and Bird 2002) , American Kestrels on SCI occasionally nested in cavities that were partially to completely obscured by vegeta- tion, typically cacti (e.g., Opuntia sp., Bergerocactus emoryii) or morning glory. We suggest that these sites could be selected to prevent predation by Common Ravens ( Corvus corax) . Ravens have been observed clinging to canyon walls attempting to ac- cess nests, and in some cases directly taking Amer- ican Kestrel chicks on SCI (B. Sullivan pers. ob- serv.). Use of prickly pear, cholla, and snake cholla to obscure nest entrances may be a form of nest defense on SCI. American Kestrels typically nest in cavities with entrances facing away from the prevailing direction of seasonal storms, or those typically oriented to the east or southeast (Balgooyen 1976, 1989, Brau- ning 1983) . West-facing nest boxes in California ex- hibited high failure rates (Bloom and Hawks 1983). We also found that American Kestrel nest cavities were oriented away from the prevailing northwest wind and toward the morning sun (Fig. 2). Nest-cavity selection may be especially impor- tant when driving wind and rain, associated with strong frontal systems, may make cavity microcli- mates on west-facing slopes undesirable. Linear Abundance. American Kestrel breeding density varies with nest site and resource availabil- ity. Density estimates range from 0.11-1.74 pairs/ km^, but can be as high as 24.7 pairs/km^ (Bird and Palmer 1988, Smallwood and Bird 2002). In Patagonia, the density of six American Kestrel nests found in basaltic cliffs was 0.038 pairs/km^ (De Lucca and Saggesse 1993). In linear habitat along the Big Lost River, southeastern Idaho, Craig and Trost (1979) reported 0.5-0. 9 nests/km of river. In South America, kestrel territories averaged 0.12/ km^ (Balgooyen 1989). Compared to the estimates in Craig and Trost (1979) and Balgooyen (1989), American Kestrels on SCI nested in higher linear abundance (2.1 pairs/km). A relatively small pro- portion of the island’s surface is made up of can- yons (ca. 9%, Institute for Wildlife Studies unpubl. data) , however, even fewer tree cavities likely exist. Thus, nesting areas confined to canyons, coupled with an adequate prey base, may combine to sup- port American Kestrel abundance to above average levels on SCI. On SCI, considerable variation in American Kes- 328 Sullivan et al. VoL. 37, No. 4 trel abundance between canyons may occur due to extreme inter-canyon variation in the amount and types of cliffs available. The subsample of eight canyons used for abundance were all located on the west side of SCI, where canyons tend to be lon- ger, shallower, and have less relief than those of the east side. Due to restricted access to the east side of SGI, a thorough census was not completed for any canyon, thus extrapolation of abundance to these areas may not be warranted. Breeding American Kestrels on SCI are a recent phenomenon (Jorgensen and Ferguson 1984). Neither Grinnell (1897) nor Mearns (cited in Jor- gensen and Ferguson 1984) detected any Ameri- can Kestrels during their initial spring visits to SCI in the late 1800s. Linton (1908) reported them as “occasional” by 1907; however, he made no men- tion of this species breeding during an April visit to SCI. Howell (1917) considered them to be “probably resident” by 1917. Jorgensen and Fer- guson (1984) counted 70 kestrels on 2 January 1980, and considered them a common resident. Currently, counts as high as 249 birds from 9-15 November 2001 (Institute for Wildlife Studies, un- publ. data) suggest a further increase in numbers. We hypothesize that the apparent change in breeding status of this species, from unknown to abundant, may be associated with the loss of native ground cover to feral herbivores, and the subse- quently ideal habitat for foraging American Kes- trels. Lavatera reportedly constituted an “unbro- ken forest, extending for miles upon the high plateaus,” and Dudleya “covered the ground throughout the entire island,” prior to the intro- duction of feral grazers (Raven 1963). With the decimation of SCFs historic vegetation, more open habitat was created, possibly allowing American Kestrels on SCI to flourish. Acknowledgments This study was funded by the Commander in Chief, Pacific Fleet, and was conducted under contract number N68711-99-C-6665, with logistical and technical support from the U.S. Navy Region Southwest, Environmental De- partment, Natural Resources Office, San Diego, CA. We thank the Natural Resources Office, Commander, Navy Region Southwest, and Southwest Division, Naval Facili- ties Engineering Command for funding this research. We thank Jonathan H. Plissner and Suellen Lynn for provid- ing useful comments on an earlier draft of this manu- script. The critiques of T.G. Balgooyen, P.B. Wood, and an anonymous reviewer improved the final version of this manuscript. Literature Cited Balgooyen, T.G. 1976. Behavior and ecology of the American Kestrel {Falco sparverius 1.) in the Sierra Ne- vada of California. Univ. Calif. Publ. Zool. 103:1-83. . 1989. Natural history of the American Kestrel in Venezuela./. Raptor Res. 23:85—93. Bird, D.M. and R.S. Palmer. 1988. American Kestrel. Pages 253-290 in R.S. Palmer [Ed.], Handbook of North American birds, Vol. 5. Diurnal raptors. Part 2. Yale Univ. Press, New Haven, CT U.S.A. Bloom, P.H. and S.J. Hawks. 1983. Nest box use and re- productive biology of the American Kestrel in Lassen County, California. Raptor Res. 17:9-14. Brauning, D. 1983. Nest site selection of the American Kestrel {Falco sparverius). Raptor Res. 17:122. Craig, T.H. and C.H. Trost. 1979. The biology and nest- ing density of breeding American Kestrels and Long- eared Owls on the Big Lost River, southeastern Idaho. Wilson Bull. 91:50-61. De Lucca, E.R. and M.D. Saggesse. 1993. Nesting of the American Kestrel {Falco sparverius) in Patagonia. Hor- nero 13:302-305. Grinnell, J. 1897. Report on the birds recorded during a visit to the islands of Santa Barbara, San Nicolas and San Clemente in the spring of 1897. Pasadena Acad. Sci. 1:4-26. Howell, A.B. 1917. Birds of the islands off the coast of southern California. Pac. Coast Avif. 12. Jorgensen, P.D. and H.L. Ferguson. 1984. The birds of San Clemente Island. West. Birds 15:111-130. Kellogg, E.M. and J.L. Kellogg. 1994. San Clemente Island vegetation condition and trend and the ele- ments of ecological restoration. Southwest Division, Naval Facilities Engineering Command, San Diego, CA U.S.A. Kovach, W.L. 1994. Oriana for Windows, Ver. 1.0. Kovach Computing Services, Pentraeth, Wales, U.K. Linton, C.B. 1908. Notes from San Clemente Island. Con- dor 10:82-86. Olmstead, F.H. 1958. Geologic reconnaissance of San Clemente Island, California. U.S. Geol. Surv. Bull. 107l-B:55-68. Raven, P.H. 1963. A flora of San Clemente Island, Cali- fornia. Aliso 5:289-347. Rohrbaugh, R.W., Jr. and R.H. Yahner. 1997. Effects of macrohabitat and microhabitat on nest-box use and nesting success of American Kestrels. Wilson Bull. 109: 410-423. Smallwood, J.A. and D.M. Bird. 2002. American Kestrel. In A. Poole and F. Gill [Eds.], the birds of North America, No. 602. The Birds of North America, Inc., Philadelphia, PA U.S.A. AND M.W. Collopy. 1991. The responses of south- eastern American Kestrels to increased availability of nesting sites in two habitats./. Raptor Res. 25:160. AND P.J. Wargo. 1997. Nest site habitat structure December 2003 Cliff-nesting American Kestrels 329 of American Kestrels in northwestern New Jersey. Bull. N.J. Acad. Sci. 42:7-10. Smith, D.G., C.R. Wilson, and H.R. Frost. 1972. The biology of the American Kestrel in central Utah. Southwest. Nat. 17:73—83. Sward, W.L. and R.H. Cohen. 1980. Plant community analysis of San Clemente Island. Report to Naval Ocean Systems Center, San Diego, CA U.S.A. U.S. Department of the Navy, Southwest Division. 2001. San Clemente integrated natural resources management plan draft final, October 2001. Tierra Data Systems, Escondido, CA U.S.A. Vedder, J.C. and D.C, Howell. 1980. Topographic evo- lution of the southern California borderland during late Cenozoic time. Pages 7-31 in The California Channel Islands: proceedings of a multidisciplinary symposium. D.M. Powers [Ed.], Santa Barbara Muse- um of Natural History, Santa Barbara, CA U.S.A. Received 4 November 2002; accepted 3 August 2003 Short Communications J Raptor Res. 37(4):330-333 © 2003 The Raptor Research Foundation, Inc. Cuban Raptor-Migration Counts in 2001 Freddy Rodriguez Santana and Luis Median Hernandez BIOECO, Museo de Historia Natural Tomas Romay, Enramada #601, CP 90 1 00, Santiago de Cuba, Cuba Mark Martell The Raptor Center, University of Minnesota, 1920 Fitch Avenue, St. Paul, MN 33108 U.S.A. Keith L. Bildstein^ Acopian Center-Haxvk Mountain Sanctuary, 410 Summer Valley Road, Orwigsburg, PA 17961 U.S.A. Key Words: Osprey; Pandion haliaetus; Cuba; Falconi- formes; La Gran Piedra; migration. Scientists and conservationists often use migration counts of raptors at traditional bottlenecks and concen- tration points to document the status of raptor popula- tions and to study their migratory behavior (e.g., Haugh 1986, Bednarz et al. 1990, Bildstein and Zalles 1995, Bild- stem 1998). Most migration watch-sites, however, are in the northern Temperate Zone, and relatively few counts have been made in the tropics (Zalles and Bildstein 2000). For example, although islands in the Caribbean basin long have been recognized as important stopover sites and wintering grounds for many North American raptors including Ospreys {Pandion haliaetus), Swallow- tailed Kites {Elanoides forficatus) , Merlins {Falco colum, bar- ms), and Peregrine Falcons {Falco peregrinus), and al- though many species of migrants remain threatened in the Caribbean Islands, no one has conducted season-long migration counts in the region. Consequently, our un- derstanding of the geography and ecology of raptor mi- gration in the area remains rudimentary, compromising our ability to protect raptors there (Poole and Agler 1987, Hoffman and Harrow 1992, Crouse and Keith 1999, Zalles and Bildstein 2000, Rodriguez et al. 2001). With this in mind, we investigated the possibility of es- tablishing one or more raptor migration watch-sites in Cuba, the largest island in the region. Here we report the results of our first season of effort (2001) at two sites, Carahatas and La Gran Piedra, in central and southeast- ern Cuba, respectively. Study Area and Methods We located our counting station at La Gran Piedra based on satellite telemetry tracking data collected by M. 1 Corresponding author’s e-mail address: bildstein® hawkmtn.org Martell as part of a continental study of Osprey migration and on the observations of F. Rodriguez Santana, which together suggested the likelihood of important move- ments of Ospreys through eastern Cuba (Martell et al. 2001, Rodriguez et al. 2001). The La Gran Piedra watch- site (20°00'41"N, 75°37'38'AV) is situated atop 1234-ni La Gran Piedra, in the Sierra Maestra range. La Gran Piedra is 18 km east of Santiago de Cuba city, Santiago de Cuba province, and 9.6 km from the seacoast. Another count- ing station, Carahatas (22°55'02"N, 80°17'5TW), was used for 4 d. Carahatas is on the Atlantic coast in Matan- zas Province, north-central Cuba. We counted migrants for 18 d at La Gran Piedra be- tween 27 August and 17 October 2001, and for 4 d at Carahatas between 18 and 21 September 2001. Counts at La Gran Piedra, which regularly began at 0800 H, never lasted for more than 7 hr due to afternoon rains. Counts at Carahatas took place from 0700-1400 H, except on 18 September, when the count finished at 1800 H. We identified all migrating raptors at the count sites using lOX binoculars. Raptors were considered migrants if they appeared on the horizon to the west and flew past the watch-site using powered or soaring flight. Data were collected using the standardized daily report of the HMANA (Hawk Migration Association of North Ameri- ca) translated into Spanish. Wind speed and temperature were recorded using a Kestrel 2000 Pocket Weather Sta- tion (Nielsen Kellerman, Chester, PA). We also recorded barometric pressure. A mechanical counter (U.S. Gov- ernment Tally Counter [Heavy Duty Model]) was used to tally the hourly count for the most numerous migrant, the Osprey, at peak hours. Results We counted 1380 raptors belonging to eight species in 95 hr of observation at the two watch-sites (Tables 1 and 2). Counts at La Gran Piedra accounted for 92% of the individuals observed. Ospreys were the most common species observed at both watch-sites, accounting for 89% {N = 93) of the migrants at Carahatas and 96% {N = 1223) of migrants at La Gran Piedra. Other migrants in- cluded Swallow-tailed Kites, Sharp-shinned Hawks {Accip- 330 December 2003 Short Communications 331 Table 1. Mean daily counts, and minimum, maximum, and mean hourly counts of raptors seen at La Gran Piedra watch-site, Santiago de Cuba Province, southeastern Cuba, during 66 hr and 19 min of observation on 18 days (27 August-17 September), in autumn 2001. Species Mean Daily Count Minimum CouNT/hr Maximum Count/ hr Mean Count/ hr Osprey 67.9 0 171 18.4 Swallow-tailed Kite 0.56 0 10 0.15 Sharp-shinned Hawk 0.05 0 1 0.20 Broad-winged Hawk 0.72 0 8 0.02 Merlin 0.39 0 1 0 Peregrine Falcon 0.28 0 2 0.08 Unidentified falcon 0.28 0 2 0.08 Unidentified raptor 0.61 0 2 0.17 iter striatus) , Broad-winged Hawks (Buteo plalyplerus) , Crested Caracaras ( Caracara cheriway ) , American Kestrels (Falco sparverius). Merlins, and Peregrine Falcons. Ospreys passed La Gran Piedra at a rate of 18.4 birds/ hr, and Carahatas at a rate of 3.3/hr. Osprey numbers at La Gran Piedra peaked at 1100-1200 H, when more than 40% of all Ospreys where counted, and the flight typically began after 0800 H. No counts occurred during the late afternoon because of precipitation. All raptors observed at this watch-site were flying from west to east, following the region’s Sierra Maestra ridges, which parallel the coast of southeastern Cuba. In Carahatas, where the flight began at about 0800 H, most Ospreys passed be- tween 0900 and 1000 H. All raptors seen migrating at this watch-site were flying from northwest to southeast along the coast. Discussion Although eight species of migrants were observed, Os- preys made up the overwhelming majority of migrants counted at both watch-sites (Tables 1 and 2). Despite the fact that Ospreys typically are considered to be broad- front migrants (Poole 1989, Poole et al. 2003), our ob- servations suggest a concentrated migration corridor for the species in southeastern Cuba. We are not certain why this is so, although most studies of Osprey migration have been in northern temperate areas and the species may behave differently in the tropics. The number of Ospreys counted at the La Gran Piedra watch-site in autumn 2001 place it among the “top” five Osprey migration watch- sites in the world (Table 3), confirming earlier sugges- tions that Cuba is along an important flyway for North American Eastern Seaboard populations of the species (Hoffman and Darrow 1992, Rodriguez et al. 2001). We observed Ospreys soaring in the thermals and mov- ing east along the Sierra Maestra mountain ridges in flocks with as many as 43 birds, much greater than the flock of 11 at Cape May, New Jersey reported by Dunne et al. (1988). All flocks with Ospreys did not include oth- er species, although one was preceded by a flock of 10 migrating Swallow-tailed Kites (Bildstein et al. 2002). This species has been reported flying with Ospreys at other sites in the Caribbean (Crouse and Keith 1999). Flying in flocks may allow individual Ospreys to locate thermals more efficiently (Kerlinger 1989). Satellite-telemetry data indicate that southbound Os- preys from North America fly from the Florida Keys to northwestern Cuba, and then along the east-west axis of Cuba before crossing the Windward Passage to Hispan- iola (Martell et al. 2001, Rodriguez et al. 2001). The rel- atively high number of Ospreys seen at La Gran Piedra lead us to conclude that this site is along the main mi- Table 2. Mean daily counts, and minimum, maximum, and mean hourly counts of raptors seen at the Carahatas watch-site, Matanzas Province, north-central Cuba, during 28 hr and 45 min of observation on four days (18-21 September), in autumn 2001. Species Mean Daily Count Minimum CoUNT/hr Maximum CoUNT/hr Mean Coun'p/ hr Osprey 23.3 0 20 3.23 Sharp-shinned Hawk 0.25 0 1 0.03 Crested Caracara 0.72 0 3 0.10 Merlin 0.25 0 1 0.03 American Kestrel 0.75 0 3 0.24 332 Short Communications VoL. 37, No. 4 Table 3. Annual mean, maximum annual count, maximum Tday count, monitoring period and migration period at the world’s four “top” Osprey migration watch-sites (Zalles and Bildstein 2000) compared with that at the La Gran Piedra watch-site, Santiago de Cuba Province, in southeastern Cuba, autumn 2001. Watch-site Mean Maximum Annual Count Maximum 1-day Count Monitoring Period Migration Period Lighthouse Point, CT, U.S.A. 1920 4040 — 1980-91 120 days Cape May Point, NJ, U.S.A. 1540 2940 308 1976-85 150 days Hilton Head Island, SC, U.S.A. 1150 1560 261 1993-96 63 days Kiptopeke, VA, U.S.A. — 5775 1053 1977-96 150 days La Gran Piedra, Santiago de Cuba, Cuba — 1223 279 2001 18 days gration corridor through the region, and may be an im- portant monitoring point for this population. All other species of raptor migrants seen at La Gran Piedra watch- site in autumn 2001 were heading east along the south- ern slope of the east-west oriented Sierra Maestra range that runs parallel to the coastline of southeastern Cuba. The Swallow-tailed Kites recorded at La Gran Piedra watch-site represented the first record of this species as a migrant in eastern Cuba (Bildstein et al. 2002). The species has been reported as an uncommon transient, and possibly a rare winter resident in Cuba by Garrido and Kirkconnell (2000). The 13 Broad-winged Hawks counted at La Gran Piedra watch-site also represent the first record of this species as a migrant in eastern Cuha. Crested Caracaras are not reported as migrants in Cuba (Garrido and Garcia 1975, Garrido and Kirkconnell 2000), although Raffaele et al. (1998) reported them as vagrants in Jamaica Bay. We considered the three cara- caras observed at the Carahatas coastal watch-site to be vagrants. All three passed the site along the seacoast in directed flight (i.e., west-to-east, straight-line flapping flight), and may have been engaged in short-distance re- gional movements as reported for this species in the United States (Clark and Wlieeler 1987). Our count results at two sites in Cuba, suggest that exploratory raptor-migration counts on Caribbean Is- lands can provide important new information regarding the routes and ecology of raptor migration in the region (Bildstein et al. 2002), as well as may provide a basis for developing long-term count efforts. Once established and working in collaboration with North American watch-sites, long-term efforts at such sites can: (1) help monitor populations of species of raptors migrating through the region, (2) study migration ecology and be- havior in the tropics and compare it with that in tem- perate regions, and (3) introduce the regional human populations to the spectacle of raptor migration, thereby improving regional understanding of the connectivity of continental North America’s and the Caribbean Island’s natural resources and increasing opportunities to achieve conservation (Bildstein and Zalles 1998). Resumen. — Se ofrecen los resultados del primer conteo de aves rapaces migratorias para Cuba y el Caribe insular. Se contaron 1380 rapaces de ocho especies en dos pun- tos de observacion en el centro y sudeste de Cuba. Las Aguilas Pescadoras {Pandion haliaetus) representaron el 89% y el 96% de las aves contadas en Carahatas y en La Gran Piedra respectivamente. Otras especies observadas durante la migracion fueron el Cernicalo {Falco sparve- rius), halconcito {Falco columbarius) , halcon peregrino {Falco peregrinus) , caraira ( Caracara plancus cheriway) y el gavilancito {Accipiter striatus) , ademas, se reportan el ga- vilan bobo {Buteo platypterus) y el gavilan cola de tijera {Elanoides forficatus) por primera vez migrando en la re- gion oriental de Cuba. Se confirma la importancia de Cuba para la migracion de la poblacion norteamericana de aguila pescadora {P. haliaetus) durante su migracion a traves del Caribe. [Traduccion de los autores] ACKNOWI.EDGMENTS Our work was made possible with support from BIOE- CO, Santiago de Cuba; Hawk Mountain Sanctuary, Kempton, Pennsylvania; and The Raptor Center, St. Paul, Minnesota. We thank all of these organizations. We also thank Eduardo Inigo-Elias, Michael McMillian, Joan Mor- rison, and an anonymous reviewer for helping improve our presentation. This is Hawk Mountain Sanctuary con- servation-science contrihution number 101. Literature Cited Bednarz, J.C., D. Klem, L.J. Goodrich, and S.E. Senner. 1990. Migration counts of raptors at Hawk Mountain, Pennsylvania, as indicators of population trends 1934-1986. Auk 107:96-109. Bildstein, K.L. 1998. Long term counts of migrating rap- tors: a role for volunteers in wildlife research./. Wildl. Manage. 62:435-445. AND J.I. Zalles (Eds.). 1995. Raptor migration watch-site manual. Hawk Mountain Sanctuary Associ- ation, Kempton, PA U.S.A. AND . 1998. Moving targets: the science and December 2003 Short Communications 333 conservation of migrating raptors in the Western Hemisphere. Torgos 28:97-108. , F. Rodriguez Santana, L.O. Median, and M. Martell. 2002. A new migration route for Swallow- tailed Kite Elanoides forficatus in east Cuba. Cotinga 18: 93-95. CiAJiK, W.S. and B.K. Wheeler. 1987. Hawks of North America. Houghton Mifflin, Boston, MA U.S.A. Crouse, D.G., Jr. and A.R. Keith. 1999. A remarkable Osprey flight and first record of Swallow-tailed Kite for Hispaniola. Pitirre 12:91. Dunne, R, D. Sibley, and C. Sutton. 1988. Hawks in flight, the flight identification of North American mi- grant raptors. Houghton Mifflin, Boston, MA U.S.A. Garrido, O.H. and F. Garcia. 1975. Catalogo de las aves de Cuba. Academia de Ciencias de Cuba, Habana, Cuba. and a. Kjrkconnell. 2000. Field guide to the birds of Cuba. Cornell Univ. Press, Ithaca, NY U.S.A. Haugh, J.R. 1986. Raptors in migration. Pages 35-48 in S.E. Senner, C.M. White, and J.R. Parrish [Eds.], Rap- tor conservation in the next 50 years. Raptor Re- search Report 5. Raptor Research Foundation and Hawk Mountain Sanctuary, Kempton, PA U.S.A. Hoffman, W. and H. Darrow. 1992. Migration of diurnal raptors from the Florida Keys into the West Indies. HMANA Hawk Migrat. Stud. 17:7-14. Kerlinger, P. 1989. Flight strategies of migrating hawks. Univ. of Chicago Press, Chicago, IL U.S.A. Martell, M.S., CJ. Henny, P.E. Nye, and MJ. Solensky. 2001. Fall migration routes, timing, and wintering sites of North American Ospreys as determined by sat- ellite telemetry. Condor 103:715-724. Poole, A. 1989. Ospreys: a natural and unnatural history. Cambridge Univ. Press, Cambridge, U.K. AND B. Agler. 1987. Recoveries of ospreys banded in United States, 1914—1984./. Wildl. Manage. 51:148- 155. Poole, A.F., R.O. Bierregaard, and M.S. Martell.. 2003 Osprey {Pandion haliaetus). In A. Poole and F. Gill [Eds.], The birds of North America, No. 683. The Academy of Natural Sciences, Philadelphia, PA, and The American Ornithologists’ Union, Washington, DG U.S.A. Raffaele, H., J. Wiley, O.H. Garrido, A. Keith, and J Raffaele. 1998. A guide to birds of the West Indies Princeton Univ. Press, Princeton, NJ U.S.A. Rodriguez, E, M. Martell, P. Nye, and K.L. Bii.dstein 2001. Osprey migration through Cuba. Pages 107-117 in K.L. Bildstein and D. Klem, Jr. [Eds.], Hawk- watching in the Americas. Hawk Migration Associa- tion of North America, North Wales, PA U.S.A. Zalles, J.I. and K.L. Bildstein (Eds.). 2000. Raptor watch: a global directory of raptor migration sites BirdLife International, Cambridge, U.K. and Hawk Mountain Sanctuary, Kempton, PA U.S.A. Received 17 October 2002; accepted 30 June 2003 Associate Editor: Joan L. Morrison J. Raptor Res. 37(4);333— 337 © 2003 The Raptor Research Foundation, Inc. Territory Change and Nest-site Switching in the Bearded Vulture ( Gypaetus barbatus) Antoni Margalida^ Bearded Vulture Study and Protection Group, Apdo. 43, E-25520, El Pont de Suert (Lleida), Spain JORDI CaNUT Generalitat de Catalunya, Departament de Medi Ambient, Parc Nacional d’Aigiiestortes i Estany de Sant Maurici, C/ Prat del Guarda 4, E-25597 Espot (Lleida), Spain Diego Garcia Generalitat de Catalunya, Departament de Medi Ambient, C/ Dr. Roux 80, E-0801 7 Barcelona, Spain Key Words: Bearded Vulture, Gypaetus barbatus; conspe- of a nesting site for a given raptor species (Newton 1979, cific interaction-, nest competition-, territory change. Collias and Collias 1984). The increase in the size of a species’ population can cause a reduction in the avail- The presence of old nests may indicate the suitability ability of suitable nesting sites and lead to intraspecific and interspecific competition for those sites. In this re- spect, some species may adopt the strategy of occupying ^ E-mail address: margalida@gauss.entorno.es 334 Short Communications VoL. 37, No. 4 or usurping other species’ nests, as has been documented in vSeveral raptors (e.g., Fernandez and Donazar 1991, Margalida and Garcia 1999). This process can lead to changes in the territory used by specific birds. In several raptor species, these changes in territories have been re- lated to the age of the birds involved to the quality of the territory, and to mate loss (Newton 1986, Wiklund 1996, Forero et al. 1999). The Bearded Vulture {Gypaetus barbatus) is a territorial, cliff-nesting vulture, whose diet mostly includes bony re- mains of medium-sized ungulates (Hiraldo et al. 1979, Donazar 1993), a resource which is unpredictable in time and space. In the Pyrenees (northeastern Spain), the number of occupied territories and the number of breed- ing pairs have shown a significant increase over the last 15 yr (annual increase of 5%; Heredia and Margalida 2001). Each pair has several nests in its territory (2—11; Heredia 1991, Margalida and Garcia 1999), that are used on a rotational basis. These nests are on average 992 m from one another (Donazar et al. 1993). The function- ality of having several nests at one territory may be to advertise the birds’ territory or to avoid the accumulation of ectoparasites by changing the nest from one year to another (Newton 1979, Margalida and Bertran 2000a). In the Bearded Vulture, both sexes participate in ter- ritorial defense (Margalida and Bertran 2000b), an activ- ity that is generally centered within a radius of 500 m from the nest used that year (Brown 1990, Bertran and Margalida 2002). Despite this territorial behavior, the Bearded Vulture has suffered from nest stealing, mainly by the Eurasian Griffon {Gyps fulvus). Also, Bearded Vul- tures have usurped the nests of other species such as the Golden Eagle (Aquila chrysaetos) and the Common Raven {Gorvus corax) (Margalida and Garcia 1999) suggesting interspecific competition for nest sites. For some species such as the Bearded Vulture, nest building involves a con- siderable investment of time and energy and, both the nest location and the material it is made of can influence breeding success (Margalida and Bertran 2000a) . On the other hand, the increase in density of nesting vultures can lead to a larger number of intraspecific interactions, a factor that can also influence this species’ breeding suc- cess (Margalida et al. 2003). Nevertheless, probably as a result of the territoriality and the relatively great distance between neighboring nests (11 km; Donazar et al. 1993), the occupation of nests by conspecifics has not been doc- umented in the Bearded Vulture. Here we document the first case of a Bearded Vulture pair’s change of territory and the first conspecific usur- pation of a neighboring pair’s nest. Study Area and Methods The study was carried out in Catalonia (northeastern Spain). In this region, the population of Bearded Vulture increased from seven territories in 1984 to 30 in 2002 (Margalida and Garcia 2002). Between 1992 and 2002, we intensively monitored 179 breeding attempts of 14- Figure 1. Location and distances between the three nesting sites occupied by the two pairs of Bearded Vul- tures. The circles indicate the location of the nests. The triangle in NS 3 indicates the nest shared by the two pairs, occupied by PA in 1996 and pair PB in 2001. Pair PA primarily used NS 2 (1997-2002) and pair PB pri- marily used NS 1 (1995-2000 and 2002). 20 focal pairs per year (Garcia et al. 1996, Margalida et al. 2003). In this area, the mean distance between nests of the same pair was 1.7 km and the mean distance be- tween nests of neighboring pairs was 12.7 km {N =18; Garcia et al. 1996). Nests were observed throughout the breeding season at least once a week. Observations began in September, coinciding with the start of nest building (Margalida and Bertran 2000a) , and ended during fledg- ing ( June-July, Margalida et al. 2003) . The identification of the birds was based on comparison of the individual characteristics of their plumage, specifically the patterns on pectoral bands (black feathers in the breast and ven- tral region) and crowns, (see Delibes et al. 1984). The long molting process during the breeding period also fa- cilitated identification of the birds during extensive pe- riods of time. Resulis Pair A: In 1996, we located this new breeding pair (PA), which occupied a vacant nesting sector (NS 3). The pair laid the eggs, incubated them, but failed during hatching. The nearest breeding pair (PB) was 11 km away in sector NS 1. After the breeding failure, PA built a sec- ond nest on the same rock face, ca. 150 m away, and it was not used again. During the pre-laying period of 1996-97, PA moved 8 km east, to an area 3 km from PB (Fig. 1). In this new sector (NS 2) PA bred in three dif- December 2003 Short Communications 335 Table 1. Summary of the breeding chronology and nest success of pairs PA and PB in the three nesting sectors studied, 1995-2002. Year Nesting Site 3 Nesting Site 2 Nesting Site 1 Pair Laying Date^^ Success- ful Repro- duction Pair Laying Date^ Success- ful Repro- duction Pair Laying Date"* Success- ful Repro- duction 1995 — — — — — — PB <8 Jan No 1996 PA <11 Jan No — — — PB No eggs laid No 1997 — — — PA ? Yes PB No eggs laid No 1998 — — — PA 12-13 Jan Yes PB 5-6 Feb No 1999 — — — PA 10 Jan Yes PB 5-7 Feb No 2000 — — — PA 7-10 Jan Yes PB 20-26 Jan Yes 2001 PB 20-26 Jan Yes PA <11 Jan Yes — — — 2002 — — — PA <8 Jan Yes PB >20 Jan No ■*Jan = January, Feb = February. ferent nests, that were built between 1996 and 2001. In 2002, PA remained in NS 2 and bred successfully. Pair B: This pair (PB) was discovered in 1984. From 1984-94 it bred successfully each year in NS 1, using five different nests, which were grouped in a 300 m radius. In 1995, PB failed during breeding and this coincided with the presence of a third individual, who was tolerated and even entered the nest along with the two members of the pair. Between 1996 and 1999, the pair failed in its breeding attempts. In 2000, PB bred successfully in NS 1 and, during the predaying period of the next breeding season (2000-01), disappeared from breeding sector NS 1. That year, PA occupied one of the nests in NS 2, while another incubating pair was located in NS 3, whose phe- nology and plumage characteristics matched that of PB (Table 1). In 2001, PB occupied a second nest in NS 3 (built by PA in 1996) and bred successfully, having moved 11 km. During 2002, PB rebuilt the first nest occupied by PA tn NS 3. PB defended the territory and reconstructed the nest up to 19 January 2002 (the date of the last obser- vation of the two individuals in NS 3). After this, these vultures were not seen again. During a later visit, an in- dividual from PB was located incubating in one of the nests in NS 1. The mean laying dates obtained between 1996 and 2002 differ significantly from one territory to another (PA: 10 January, PB: 28 January; Table 1; Mann-Whitney U-test, z = — 2.V4, P = 0.0062). Plumage characteristics of the individuals and egg-laying dates (the latter being very regular each year for all pairs; Margalida et al. 2003) corroborate the fact that NS 3 was shared in different years by PA and PB (Table 1). Discussion In the study area, the mean distance from one nest to the next nearest nest in the same territory is 1.7 km, a greater distance than the one obtained by Donazar et al. (1993) in the central Pyrenees (0.99 km). This fact may facilitate nest “stealing” and force the Bearded Vulture to look for other sites to breed (Margalida and Garcia 1999). The territory change observed (moving 11 km) could be a result of nests being occupied by heterospe- cifics. However, PB had all its nests and had stayed in the same territory for 18 years. In many bird species, the de- cision to move can be based on the individual’s current breeding success or the quality of the territory (Switzer 1997). On the other hand, in several raptor species, mate loss and lack of breeding success are related to territory changes (Newton and Marquiss 1982, Wiklund 1996, For- ero et al. 1999). The change in egg-laying dates (delayed 1 mo) in PB suggests that at least one of the members of the pair was replaced (probably the female), because egg-laying dates are usually very regular from one year to the next in the Bearded Vulture, with a year to year var- iability of ±4.2 d (Margalida et al. 2003). Because the replacement must have taken place 5—6 yr before the ter- ritory change, this fact does not appear to explain the territory switch. Moreover, after the successful breeding in NS 3, PB returned to its original territory (NS 1) and this behavior has been observed in other pairs in which there have apparently been no replacement breeders (e.g., in territories with nests usurped by Eurasian Grif- fons) . Another factor, which might explain the move may be related to human disturbance, which can disrupt this spe- cies’ breeding success (Donazar et al. 1993, Margalida et al. 2003) . Because human pressure is scarce in the three study sectors (pers. observ.), this factor does not appear to explain the move. Moreover, after success for breeding in NS 3, PB went back to its old breeding sector (NS 1) Finally, the increases in the density of birds can cause interactions related to competition for the exploitation of trophic resources and nesting sites (Margalida and 336 Short Communications VoL. 37, No. 4 Bertran 2003), which lead to the territory change. Nev- ertheless, during the monitoring of the pairs studied, no agonistic interactions were observed. In addition, in the study area there are pairs that coexist at lesser distances and they breed successfully and have not changed terri- tory (Margalida and Garcia 2002, pers. observ.). None of the hypotheses consider appear to explain the observed nest change by Bearded Vultures. Although the specific reasons for the move are unknown, one plausible explanation would be related to the availability of trophic resources and the optimization of the time and energy this would have meant for the pair in order to exploit them. The effect of breeding success and mate loss on territory change could be a consequence of a response to local food conditions (Newton and Marquiss 1982). The presence of predictable sources of food, whether these be feeding stations or areas where farmers leave carcasses, may be favoring the settlement of new pairs (Heredia and Razin 1999, A. Margalida pers. observ.). The search for, preparation, and transportation, of food represent a significant part of the parental effort the spe- cies invests during breeding (Margalida and Bertran 2000b, Margalida and Bertran 2001). The extensive for- aging areas usually exploited by a pair of Bearded Vul- tures measure over 600 km^ and lineal movements of over 30 km can be common (pers. observ.). PB regularly exploited areas closer to NS 3 than NS 1 as a result of the abundance of Pyrenean Chamois {Rupicapra pyrenai- ca), a very important trophic resource for breeding pairs m the Pyrenees (Margalida et al. 1997, Garcia and Mar- galida 1998). Movements during foraging should allow the pair to learn of potential nesting sites as well as the seasonal changes in the availability of trophic resources. A seasonal and temporary abundance in food resources may have caused this move, with the benefits (proximity of abundant food and quality territory indicated by the presence of conspecihc nests) outweighing the costs of temporarily abandoning the original territory (loss of the territory and the nests due to conspecilics or heterospe- cifics occupying them). Nevertheless, more information IS needed on territory change in this species to explain the proximate factors stimulating such action. Resumen. — Documentamos el cambio de sector de nidi- ficacion en una pareja de quebrantahuesos {Gypaetus bar- batus) y el primer caso de ocupacion de un nido de la misma especie. En 1996 una pareja (A) colonizo el sector de nidihcacion NS 3. La pareja vecina mas proxima (B) se encontraba all km en NS 1. En 1997 la pareja A se tra.slado a NS 2, situandose a tan solo 3 km de NS 1. Tras mdificar 18 ahos en NS 1, en el aho 2000 la pareja B se desplazo a NS 3 para ocupar el nido reconstruido por la pareja A en 1996, criando con exito. Finalmente en 2001 la pareja B volvio a NS 1 quedando vacante de nuevo NS 3 Aunque se desconocen las causas de dicho desplaza- miento, el cambio de sector de nidihcacion podria estar relacionado con la posible abundancia temporal de los recursos trohcos y la optimizacion en la explotacion de estos, ya que NS 3 se encuentra en el area de campeo habitual de la pareja B. [Traduccion de los autores] Acknowledgments J. Bertran, J.A. Donazar, andj. Vinuela provided help- ful comments on an earlier version of this manuscript. S. Hardie translated the text into English. This study was supported by Departament de Medi Ambient de la Ge- neralitat de Catalunya. Literature Cited Bertran, J. and A. Margalida. 2002. Territorial behav- iour of the Bearded Vulture ( Gypaetus barbatus) in re- sponse to the Griffon Vulture {Gyps fulvus). J. Field Ornithol. 73:86-90. Brown, C.J. 1990. Breeding biology of the Bearded Vul- ture in southern Africa, Parts I-III. Ostrich 61:24-49. Collias, N.E. and E.C. Collias. 1984. Nest building and bird behaviour. Princeton Univ. Press, Princeton, NJ U.S.A. Delibes, M., F. Hiraldo, and J. Calderon. 1984. Age and geographic variation in the Bearded Vulture Gypaetus barbatus (Linnaeus, 1758). Bonn. Zool. Beitr. 35:71-90. Donazar, J. A. 1993. Los buitres ibericos. Biologfa y con- servacion. J.M. Reyero, Madrid, Spain. , F. Hiraldo, and J. Bustamante. 1993. Factors influencing nest site selection, breeding density, and breeding success in the Bearded Vulture ( Gypaetus bar- batus). J. Appl. Ecol. 30:504—514. FernAndez, C. andJ.A. Donazar. 1991. Griffon Vultures {Gyps fulvus) occupying eyries of other cliff-nesting raptors. Bird Study 38:42-44. Forero, M.G., J.A. DonAzar, J. Bias, and F. Hiraldo 1999. Causes and consequences of territory change and breeding dispersal distance in the Black Kite. Ecol- ogy 80:1298-1310. GarcIa, D., a. Margalida, X. Parellada, and J. Canut. 1996. Evolucion y parametros reproductores del Que- brantahuesos Gypaetus barbatus en Catalunya (NE Es- paha) . Alauda 64:227-236. AND . 1998. Avaluacio de la disponibilitat trohca per al Trencalos {Gypaetus barbatus) al Parc Na- cional d’Aiguestortes i Estany de Sant Maurici: im- portancia de les poblacions d’Isards {Rupicapra pyre- naica). Pages 179-188 in IV Jornades sobre Recerca al Parc Nacional d’Aiguestortes. Generalitat de Cata- lunya, DARP, Lleida, Spain. Heredia, R. 1991. Biologia de la reproduccion. Pages 27- 38 in R. Heredia and B. Heredia [Eds.], El Quebran- tahuesos {Gypaetus barbatus) en los Pirineos. ICONA, Coleccion Tecnica, Madrid, Spain. AND A. Margalida. 2001. Status, breeding param- eters and conservation measures in the Spanish Bearded Vulture {Gypaetus barbatus) population. Pages 51-57 in K. Sakoulis, M. Probonas, and S. Xirouchakis December 2003 Short Communications 337 [Eds.], Proceedings of the 4th Workshop of Bearded Vulture. Natural History Museum of Crete and Uni- versity of Crete, Iraklio, Greece. AND M. Razin. 1999. Ecology and conservation of the Bearded Vultures: the case of the Spanish and Erench Pyrenees. Pages 29-45 in M. Mylonas [Ed.], Proceedings of the Bearded Vulture Workshop. Nat- ural History Museum of Crete and University of Crete, Iraklio, Greece. Hiraldo, R, M. Delibes, and J. Calderon. 1979. El Que- brantahuesos {Gypaetus barbatus). L. Monografias 22. ICONA, Madrid, Spain. Margauda, a. and J. Bertran. 2000a. Nest-building be- haviour of the Bearded Vulture {Gypaetus barbatus). Ardea 88:259-264. and . 2000b. Breeding behaviour of the Bearded Vulture ( Gypaetus barbatus) : minimal sexual differences in parental activities. Ibis 142:225-234. AND . 2001. Function and temporal varia- tion in the use of ossuaries by Bearded Vultures {Gypaetus barbatus) during the nestling period. Auk 118:785-789. AND . 2003. Interspecific and intraspecific kleptoparasitic interactions of the Bearded Vulture {Gypaetus barbatus) at nesting areas./. Raptor Res. 37: 157-160. AND D. Garcia. 1999. Nest use, interspecific rela- tionships, and competition for nests in the Bearded Vulture {Gypaetus barbatus) in the Pyrenees: influence on breeding success. Bird Study 46:224-229. AND . 2002. Pla de recuperacio del trenca- 16s a Catalunya: biologia i conservacio. Documents dels Quaderns de Medi Ambient, 7. Generalitat de Catalunya, Departament de Medi Ambient, Barcelo- na, Spain. , , J. Bertran, and R. Heredia. 2003. Breed- ing biology and success of the Bearded Vulture {Gypaetus barbatus) in the eastern Pyrenees. Ibis 145 244-252. , , and R. Heredia. 1997. Estimacion de la disponibilidad trofica para el Quebrantahuesos en Ca- taluna (NE Espana) e implicaciones sobre su conser- vaci6n. Donana, Acta Vertebr. 24:227-235. Newton, I. 1979. Population ecology of raptors. T. & A D Poyser, Berkhamsted, UK. . 1986. The Sparrowhawk. T. 8c A.D. Poyser, Cal- ton, UK. and M. Marquiss. 1982. Fidelity to breeding area and mate in Sparrowhawk Accipiter nisus. J. Anim. Ecol 51:327-341. Switzer, P.V. 1997. Past reproductive success affects fu- ture habitat selection. Behav. Ecol. Soc. 40:307-312 WiKLUND, C.G. 1996. Determinants of dispersal in breed- ing Merlins {Ealco columbarius) . Ecology 77:1920-1927. Received 31 January 2003; accepted 11 August 2003 J. Raptor Res. 37(4):337-340 © 2003 The Raptor Research Foundation, Inc. Immature Northern Goshawk Captures, Kills, and Feeds on Adult-sized Wild Turkey Gregory H. Golet^ The Nature Conservancy, 500 Main Street, Chico, CA 95928 U.S.A. Henry T. Golet 5-1 Binny Road, Old Lyme, CT 06371 U.S.A. Albie M. Colton 19 Burr Road, Lyme, CT 06371 U.S.A. Key Words: Northern Goshawk; Accipiter gentilis; Wild Turkey; Meleagris gallopavo; food-niche breadth; predator-prey interaction; sexual size dimorphism. The largest prey regularly taken by Northern Goshawks ^ E-mail address: ggolet@tnc.org {Accipiter gentilis) are snowshoe hares {Lepus americanus) (Squires and Reynolds 1997). Although remains of Wild Turkey {Meleagris gallopavo) polts have been reported in goshawk pellets (Bosakowski et al. 1992), we are not aware of any record of a Northern Goshawk killing a full- grown turkey. Here, we document an observation of an immature goshawk killing and feeding on a full-grown juvenile Wild Turkey in Connecticut. 338 Short Communications VoL. 37, No. 4 Figure 1. A female Northern Goshawk {Accipiter gentilis) stands above a yearling Wild Turkey {Mekagris gallopavo) that it killed the previous day at Lyme, Connecticut. Pho- tograph (by H. Golet) taken with a Kodak 240 digital camera through a Bushnell Spacemaster spotting scope (20 X eyepiece) from ca. 10 meters away. Methods The initial attack sequence was observed (by A. Col- ton) at close range from inside the house. All subsequent observations were made outside with binoculars and a spotting scope (ca. 30 m from the carcass) aimed through a gap in a mountain laurel {Kalmia latifolia) thicket. Continuous observation periods were confined to the early mornings of each day, with additional checks on the carcass being opportunistic in nature. It is likely, therefore, that some feeding bouts were missed. Results On 12 March 2002, at about 0930 H a Northern Gos- hawk {Accipiter gentilis) attacked and killed a Wild Turkey {Mekagris gallopavo) at Lyme, Connecticut. The goshawk struck the turkey while it was feeding alone in a small clearing (3 m X 12 m) beneath a backyard bird feeder situated ca. 5 m from a house. The turkey was unaware of the goshawk until the moment of the attack, when a strike on the back drove it to the ground, causing an explosion of feathers. Immediately the turkey rose and ran, head down, toward a nearby (2 m away) mountain laurel thicket that forms the border of the clearing and the adjoining oak/hickory {Quercus/ Carya) forest. The hawk “rode” the turkey to the edge of the clearing, but then jumped off and gave pursuit by flight. The goshawk pumped its wings rapidly while flying within 1 m of the ground. Approximately 50 m from the location of the original attack the goshawk overtook the turkey, again pouncing upon its back. Meanwhile, a small flock of American Crows {Corvus brachyrhynchos) congregated overhead, flying in a tight circle and calling loudly. Pinned to the ground breast down, the turkey beat its wings frantically while the hawk used a kneading action. Figure 2. The yearling Wild Turkey carcass as it ap- peared after the first day of being fed upon by the gos- hawk. Photograph by H. Golet. repeatedly bending over and straightening up, to drive its talons into its prey. Less than 1 min elapsed from the time the hawk initially struck the turkey until it ceased flapping, unconscious on the forest floor. Shortly after the completion of the kill, the goshawk and nearby crows flew off, perhaps startled by our presence. Within 1 hr the goshawk returned to feed on the tur- key. It crouched virtually motionless for the first 5 min following its return, although at one point it spread its wings (mantling) briefly to obscure the carcass from the view of a Turkey Vulture {Cathartes aura) that flew over- head. The goshawk then began to pluck and eat the tur- key, which it did for 20 min before flying off. The hawk made no attempt to conceal the turkey by covering it with leaves prior to its departure, cilthough caching behavior has been observed in goshawks previously (Schnell 1958). The hawk fed on the carcass sporadically over the next three days (Fig. 1), primarily in the early morning, but occasionally at midday and in the early afternoon. On 14 March (day 3) it arrived before 0610 H, and fed for over 1 hr. The hawk consumed much of the dorsal muscula- ture (primarily the iliotibialis and latissimus dorsi) on the back of the turkey, partially exposing the bones of the sternum and pelvic girdle (Fig. 2). At 0615 H on 15 March (day 4) , the turkey carcass was found ca. 3 m from its original position. The hawk did not return to the car- cass that morning (through 0830 H at least), although it was observed feeding on it at around 1300 H, Sometime after this observation, and before 0600 H on 16 March (day 5), the carcass disappeared from the site, although there was no evidence of it being dragged through the leaves as was the case the day before. Discussion Our observation suggests that prey biomass and food- niche breadth may be larger for goshawks than has been December 2003 Short Communications 339 previously recognized. We know of no previous record of a goshawk killing a full-grown Wild Turkey, although the remains of turkey poults have been found in goshawk pellets in the New Jersey-New York Highlands (Bosa- kowski et al. 1992) and in prey remains on the Kaibab Plateau, Arizona (R. Reynolds pers. comm.). The imma- ture goshawk, which we suggest was a female based on a consultation with R. Reynolds (pers. comm.), likely weighed ca. 1005 g (Mueller et al. lOVO), or ca. 4X as much as the yearling female turkey (ca. 3900 g, Eaton 1992). This prey differential is 63% larger than what has been previously recorded for female goshawks (2.4X their mass) based on observations of snowshoe hare {Le- pus americanus) predation (Squires and Reynolds 1997). The turkey weighed ca. lOX the previously calculated mean goshawk prey masses of 307 g in Oregon (Reynolds and Meslow 1984), 271 g in New Jersey (Bosakowski et al. 1992), 303 g in Connecticut (Bosakowski et al. 1992), and 413 g in Washington (Watson et al. 1998). Also to our knowledge, this is the first record of a gos- hawk returning to a prey item over such an extended period (four successive days) , although a mule deer ( Odo- coileus hemionus) gut pile left by hunters in Wyoming was visited by a goshawk on two subsequent days (Squires 1995). The turkey may have been easier to kill because it was a yearling. Goshawks frequently catch young animals (Opdam et al. 1977), which may be more vulnerable to predation than adults, especially in late winter (Cresswell and Whitfield 1994). Although we cannot be certain that being alone increased the susceptibility of the turkey to attack, this is likely, as group living is advantageous in terms of predator avoidance (Pulliam and Caraco 1984). Adult turkeys are typically preyed upon by mammalian predators (primarily coyotes, [Canis latrans]-, Wright et al. 1996, Hubbard et al. 1999), although Great Horned Owls {Bubo virginianus; Hubbard et al. 1999), have been ob- served to take turkey hens from night roosts (Wright et al. 1996). Other raptors, including Golden Eagles {Aquila chrysaetos] Bent 1937) and Barred Owls {Strix varia; Van- gilder and Kurzejeski 1995) have also been reported to prey on turkey hens, abeit infrequently. Goshawks are sexually dimorphic with females typically weighing 20-40% more than males (Squires and Reyn- olds 1997). This leads to the predictions that females should have wider food niche breadth, and be better buffered against fluctuations in prey availability than males (Optimality Theory; Schoener 1971). Our obser- vation, although of only one prey capture, suggests that female goshawks may indeed have a wider range of prey availability. This contrasts with previous studies conduct- ed during the breeding season that found similar prey sizes among the sexes (Snyder and Wiley 1976, Widen 1989, Boal and Mannan 1996), but is in accord with a recent radiotelemetry study conducted during winter in northern Einland. In the winter study, Tornberg and Col- paert (2001) found that as forest grouse (Tetraonidae) availability declined, female goshawks switched to preying upon hares, although males did not appear to have this option. Periods of reduced prey availability, such as win- ter (the season of our observation), may best reveal limits of prey capture that are imposed by body size. Resumen. — En marzo 12 de 2002, cerca de las 0930 H un azor norteno {Accipiter gentilis) ataco, mato y segui- damente se alimento de un pavo silvestre {Meleagris gallo- pavd) de primer ano, mientras este se alimentaba solo en un pequeno claro (3 m H 12 m) debajo de un alimen- tador para aves de jardin situado aproximadamente a 5 m de la casa. Aunque se conoce que los azores depredan sobre pavipollos, para nuestro conocimiento, este es el primer registro de que un primal sea atacado. Este en- cuentro es notable dada la gran diferencia en tamano entre el depredador y la presa. Se estimo que el pavo era 4 veces mas pesado que el azor, y aun asi este fue so- metido con relativa facilidad. El gavilan se alimento del cadaver sucesivamente durante 4 dias. Las interacciones entre estas dos especies en Nueva Inglaterra pueden ser mas comunes ahora que en las decadas pasadas dado que las poblaciones tanto de pavos como de azores parecen haber aumentado. [Traduccion de Cesar Marquez] Acknowi.edgments We thank Stephen W. Eaton, William F. Porter and Ri- chard T. Reynolds for helping identify the sex and age of the birds. Our manuscript was improved by helpful comments from J. Bednarz, E. McClaren, R.T. Reynolds, andj. Squires. Literature Cited Bent, A.C. 1937. Life histories of North American birds of prey, Part 1. U.S. Natl. Mus. Bull. 167. Boai,, C.W. and R.W. Mannan. 1996. Prey sizes of male and female Northern Goshawks. Southwest. Nat. 41; 355-358. Bosakowski, T, D.G. Smith, and R. Speiser. 1992. Niche overlap of two sympatric-nesting hawks Accipiter spp. in the New York-New Jersey Highlands. Ecography 15: 358-372. Cresswell, W. and D.P. Whitfield. 1994. The effects of raptor predation on wintering wader populations at the Tyninghame Estuary, southeast Scotland. Ibis 136: 223-232. Eaton, S.W. 1992. Wild Turkey {Meleagris gallopavo). In A. Poole, P. Stettenheim, and F. Gill [Eds.], The Birds of North America, No. 22. The Academy of Natural Sciences, Philadelphia, PA, and The American Orni- thologists’ Union, Washington, DC U.S.A. Hubbard, M.W., D.L. Garner, and E.E. Klaas. 1999. Fac- tors influencing Wild Turkey hen survival in south- central Iowa. J. Wildl. Manage. 63:731—738. Mueller H.C., D.D. Berger, and G. Allez. 1976. Age and sex variation in the size of goshawks. Bird-Banding 47:310-318. 340 Short Communications VoL. 37, No. 4 Opdam P., J. Thiessen, P. Verschuren, and G. Muskens. 1977. Feeding ecology of a population of goshawk Ac- cipiter gentilis. J. Ornithol. 118:35-51. Pulliam, H.R. and T. Caraco. 1984. Living in groups: is there an optimal group size? Pages 122-147 m J.R. Krebs and N.B. Davies [Eds.], Behavioural Ecology: an evolutionary approach, 2nd ed. Blackwell Science, Oxford, U.K. Reynolds, R.T. and C. Meslow. 1984. Partitioning of food niche characteristics of coexisting Accipiter dur- ing breeding. Auk 101:761-779. ScHNELL, J.H. 1958. Nesting behavior and food habits of goshawks in the Sierra Nevada of California. Condor 60:377-403. ScHOENER, T.W. 1971. Theory of feeding strategies. Ann. Rev. Ecol Syst. 2:369-404. Snyder, N.F.R. and J.W. Wiley. 1976. Sexual size dimor- phism in hawks and owls of North America. Ornithol. Monogr. 20. Squires, J.R. 1995. Carrion use by Northern Goshawks. J. Raptor Res. 29:283. AND R.T. Reynolds. 1997. Northern Goshawk (4c- cipiter gentilis). In A. Poole and F. Gill [Eds.], The Birds of North America, No. 298. The Academy of Natural Sciences, Philadelphia, PA, and The Ameri- can Ornithologists’ Union, Washington, DC U.S.A. Tornberg, R. AND A. Col.paert. 2001. Survival, ranging, habitat choice, and diet of the Northern Goshawk Ac- cipiter gentilis during winter in northern Finland. Ibis 43:41-50. Vangilder, L.D. and E.W. Kurzejeski. 1995. Population ecology of the eastern Wild Turkey in northern Mis- souri. Wildl. Monogr. 130. Watson, J.W., D.W. Hays, S.P Finn, and P. Meehan-Mar- TIN. 1998. Prey of breeding Northern Goshawks in Washington./. Raptor Res. 32:297-305. Widen, P. 1989. The hunting habitats of goshawks Accip- iter gentilis in boreal forests of central Sweden. Ibis 131: 205-213. Wright, R.G., R.N. Paisley, and J.F. Kubisiak. 1996. Sur- vival of Wild Turkey hens in southwestern Wisconsin. J. Wildl. Manage. 60:313-320. Received 2 December 2002; accepted 25 August 2003 J Raptor Res. 37(4):340-342 © 2003 The Raptor Research Foundation, Inc. A Helper at the Nest of Peregrine Falcons in Northern Japan Takashi Kurosawa and Reiko Kurosawa^ 3-1-6 Kojiniacho, Chofu, Tokyo 182-0026 Japan Key Words: Peregrine Falcon; Falco peregrinus japonen- sis; feeding, helper, juvenile; territory. Peregrine Falcons {Falco peregrinus) are widely distrib- uted throughout the world, and primarily breed in mo- nogamous pairs that display aggressive territorial behav- ior around their nest sites (Cade 1982). Despite numerous studies of this species in Europe, North Amer- ica, and elsewhere (e.g.. Cade et al. 1988), information on the ecology of Peregrine Falcons {F. p. japonemis) in East Asia is very limited. Cooperative breeding is infre- quent in this species (Skutch 1987) with reported excep- tions in North America and in France (Spofford 1969, Monneret 1983). Here, we describe an observation of helping at the nest of a Peregrine Falcon in Hokkaido, northern Japan. ’ Corresponding author’s e-mail address: rkurosawa® nifty.com Study Area and Methods We recorded observations on the behavior and the breeding status of Peregrine Falcons for about 120 hr each year from 1993-2002 at a study site on the Etomo Peninsula in Muroran Hokkaido, northern Japan (42°19'N, 140“59'E). Six pairs of non-migratory Pere- grines (pairs B to G) occur on a 10-km stretch of vertical cliffs, part of which is more than 100 m in height, along the narrow Etomo Peninsula (1. 0-3.5 km in width; Ku- magai 1989, Ueta et al. 1995). The six sites fledged 1.1 young (±1.1 SD) per pair per year from 1993-2002. This peninsula is also a major landfall and point of departure for migrating songbirds and raptors, and in 1998 the banding station on the peninsula recorded 57 species of small- to medium-sized land birds (T. Banno pers. comm.), which are suitable prey for Peregrines in Japan (Yamada 2002). We used 20 X binoculars and a 77x spotting scope to make observations. Because no peregrines in Japan have been marked, we attempted to identify individual birds by their characteristic features such as malar patches, ven- tral marks (Enderson and Craig 1988), favorite look-out perches and behavior toward the observers. December 2003 Short Communications 341 Results and Discussion The territorial pairs in the study area usually showed aggression toward intruding peregrines that flew <5 m above, or close to (<100 m) the nesting cliffs. On 17 March 2001, however, we observed a Peregrine with ver- tically striped breast and belly perched on a crossarm of a power pylon about 3 m from an adult male in the ter- ritory of the “F” pair. The former was obviously larger than the adult male. The adult male showed no sign of aggression toward this juvenile. About 5 min later, the juvenile peregrine departed to chase a feral rock dove {Columba livid), but came back without the prey and probably perched out of sight on the cliff, ca. 100 m from the male. An adult pair had been seen together on the wing at this site two days earlier, on 15 March. Moreover, later on 17 March we observed a copulation between the adult pair on the nesting cliff. On 24 May 2001, a peregrine with juvenile plumage carried a small prey item into the nest of the same pair. There were three nestlings with white down in the nest; they appeared to be about 2-wk old. The juvenile fed the nestlings for about 10 min and then flew around the point to the other side of the cliff. During this feeding bout, the adult pair were out of sight and neither called or came to chase away the juvenile. The territorial adult female glided by the nesting cliff about 20 min after this feeding. The clutch of three young fledged successfully in midjune, when this juvenile helper was no longer ob- served at this site. Peregrine Falcons have a broad range of post-nesting dependency periods, and particularly, non-migratory populations of falcons are reported to have prolonged juvenile-dependency periods (Sherrod 1983). In our study area, some juvenile peregrines were observed to stay in the natal territories long after the breeding sea- son; five cases recorded in October (ca. 4 mo after fledg- ing), one in December, and two in March. We assumed that these juveniles were the offspring of the territory owners for several reasons. First, they provoked no ag- gression from the territorial pairs. Secondly, no tolerated juveniles were observed in the territories of the pairs that failed to raise broods that year or the previous year. Similar to our observations, Monneret (1983) reported the helping behavior of juvenile peregrines at nests of adult pairs in the 1970s and 1980s. He suggested that juveniles were tolerated in their natal territories well into the breeding season, through the period when eggs and nestlings were present. The appearance of chicks prob- ably triggered brooding or feeding behavior by the ju- veniles, inducing them to become helpers at the nests. Monneret emphasized the importance of the kinship of the adult pair to the third falcon for the development of helping behavior to occur, because the parents would tol- erate their own young more easily than strangers in their territory. Delayed dispersal is a common route to the evo- lution of cooperative breeding in many birds including raptors (Kimball et al. 2003). Recent reviews on cooperative breeding suggest that both ecological constraints and life history traits such as low adult mortality and low dispersal rate are important factors in the evolution of cooperative breeding (Arnold and Owens 1999, Hatchwell and Komdeur 2000). Pere- grines nest on cliffs over much of their range, which re- stricts their breeding distribution (Newton 1988). They are relatively long-living with low adult mortality (Rat- cliffe 1993). Also this species is known to be philopatric (Newton and Mearns 1988). Therefore, Peregrine Fal- cons in Muroran Hokkaido may provide an interesting example of an early stage in the evolution of a coopera- tive breeding system. This sporadically-expressed behav- ior should be monitored and documented in other per- egrine studies and observers should look for factors that correlate with and may foster this breeding system. Resumen. — Lo halcones peregrinos {Falco peregrinus) usualmente procrean en parejas monogamas que des- pliegan un comportamiento territorial agresivo alrede- dor de sus sitios nido. A pesar de su territorialidad y agre- sion hacia los intrusos, una hembra juvenil fue observada en el territorio de una pareja adulta durante la estacion reproductiva en un sitio de anidacion del halcon pere- grino (K p. japonensis) en Muroran, Hokkaido, norte del Japon. Esta trajo comida y alimento a los polluelos de la pareja que estaba anidando sin que fuera expulsada. Adi- cionalmente a este individuo, otros juveniles fueron to- lerados en sus territorios natales largo tiempo despues de la estacion reproductiva, lo cual puede ocasionalmen- te conducir a que cooperen en el nido, en la region de Hokkaido, Esta especie en el norte del Japon puede ser un ejemplo interesante de un temprano estado en la evo- lucion de la reproduccion cooperativa en rapaces. [Traduccion de Cesar Marquez] Acknowled gments We would like to thank the Ministry of the Environ- ment for supporting our study. We would also like to ex- press our appreciation to S. Honda, President of Muro- ran Chapter of the Wild Bird Society of Japan, and T Banno for information on birds migrating through the study area. Special thanks goes to M. Kumagai, wildlife photographer, for providing ecological information on the peregrines of the Etomo Peninsula. We are very grateful to J. Pagel, R.W. Nelson, C. White, J.C. Bednarz, and R.A. Askins for their helpful comments and sugges- tions on earlier drafts of this manuscript. Literature Cited Arnold, K.E. and I.P.F. Owens. 1999. Cooperative breed- ing in birds: the role of ecology. Behav. Ecol. 10:465- 471. Cade, T.J. 1982. The falcons of the world. Cornell Univ. Press, Ithaca, NY U.S.A. , J.H. Enderson, C.G. Thelander, and C.M. White. 1988. Peregrine Falcon populations: their 342 Short Communications VoL. 37, No. 4 management and recovery. The Peregrine Fund, Boi- se, ID U.S.A. Enderson, J.H. and G.R. Craig. 1988. Population turn- over in Colorado Peregrines. Pages 685-688 in T.J. Cade, J.H. Enderson, C.C. Thelander, and C.M. White [Eds.], Peregrine Falcon populations: their manage- ment and recovery. The Peregrine Fund, Boise, ID U.S.A. Hatchwell, BJ. AND J.Komdeur. 2000. Ecological con- straints, life history traits and the evolution of coop- erative breeding. Anim. Behav. 59:1079-1086. Kimball, R.T., PC. Parker, and J.C. Bednarz. 2003. The occurrence and evolution of cooperative breeding among the diurnal raptors (Accipitridae and Falco- nidae). Auk 120:717-729. Kumagai, M. 1989. “Etomo hanto ni ikiru Hayabusa ta- chi” (Peregrines on the Etomo Peninsula). Anima 196:59-64. Monneret, R-J- 1983. L’aide a I’elevage chez le Faucon pelerin. Alauda 4:241—250. Newton, I. 1988. Commentary — population regulation in Peregrines: an overview. Pages 761-770 in TJ. Cade, J.H. Enderson, C.C. Thelander, and C.M. White [Eds.], Peregrine Ealcon populations: their manage- ment and recovery. The Peregrine Fund, Boise, ID U.S.A. Newton, I. and R. Mearns. 1988. Population ecology of Peregrines in south Scotland. Pages 651-665 in T.J. Cade, J.H. Enderson, C.C. Thelander and C.M. White [Eds.], Peregrine Falcon populations: their manage- ment and recovery. The Peregrine Fund, Boise, ID U.S.A. Ratcliffe, D.A. 1993. The Peregrine Falcon, 2nd Ed. T. & A.D. Poyser, London, U.K. Sherrod, S.K. 1983. Behavior of fledgling Peregrines. The Peregrine Fund, Ithaca, NY U.S.A. Skutch, A.F. 1987. Helpers at birds’ nests. Univ. of Iowa Press, Iowa City, lA U.S.A. Spofford, W.R. 1969. Extra female at a nesting site. Pages 418-419 inj. Hickey, [Ed.], Peregrine Falcon popu- lations: their biology and decline. The Univ. of Wis- consin Press, Madison, WI U.S.A. Ueta, M., R. Kurosawa, and T. Kurosawa. 1995. The home ranges of Peregrine Falcons in the Etomo Pen- insula, Hokkaido. Pages 83-89 in “Kisho yasei dous- hokubutushu seisokujokyo chosa hokokusho” (Re- ports on the status of rare wildlife: Steller’s Sea Eagle, Coshawk, and Peregrine Falcon). The Environment Agency, Tokyo, Japan. Yamada, I. 2002. Status of Peregrine Falcons Falco peregri- nus in the Setouchi District, western Japan. Goshawk 3:4-8. Received 7 January 2003; accepted 17 August 2003 J Raptor Res. 37(4) :342-344 © 2003 The Raptor Research Foundation, Inc. Two Cases of Cooperative Breeding in Eurasian Hobbies Inigo Zuberogoitia Lab. Zoologia, Dpto Zoologia Y Dindmica Celular Animal, Universidad Del Pais Vasco, APDO 644, 48080 Bilbao, Spain Jose Antonio Martinez Juan de la Cierva 43, E-03560 El Campello, Alicante, Spain Ainara Azkona, Agurtzane Iraeta, Inaki Castillo, Raul Alonso, and Sonia Hidalgo Sociedad para el Estudio de las Aves Rapaces, C/ Karl Marx 15, 4° E, 48950 Erandio, Bizkaia, Spain Key Words: Eurasian Hobby, Falco subbuteo; cooperative behavior, polyandry; polygamy. Polygamy is a common mating system in birds (Alcock 1993), and it can involve either polygyny (a male breed- ing with two or more females) or polyandry (a female ^ E-mail address: Inigo.zuberogoitia@wanadoo.es breeding with two or more males). A recent review of literature shows that group breeding has been docu- mented in 42 species of diurnal raptors (Kimball et al. 2003). Furthermore, these authors suggest that cooper- ative breeding is more common in terms of the number of species and in frequency than the available data indi- cate. In addition, they reported that 21 species of raptors showed polyandrous behavior and four other species may be polyandrous. December 2003 Short Communications 343 Polyandry may occur by extra-pair copulations (EPCs) or be a result of two or more males breeding and main- taining a bond with one female. The first case is not com- monly known in most species of raptors, in part, because the reported frequencies of mixed paternity based on DNA fingerprinting analyses are scanty (e.g., Korpimaki et al. 1996, Rohner 1997, Marks et al. 1999). The second case involves cooperative breeding behavior between two or more males and one female. Males may be related or not and the number of males that copulate with a female varies between species and sometimes only one male of the group copulates (Eamball et al. 2003). Polygamy has been rarely documented in Eurasian Hobbies {Falco subbuteo). Chapman (1999) found that a stray third bird, which could have been a yearling, would consort occasionally with an established pair. Bijlsma (1980) found that out of 26 migratory arrivals of hobbies in the Netherlands, seven involved single birds, 14 in- volved pairs, and five a “trio” of hobbies. It is unknown whether these trios became cooperatively breeding groups or not. Cramp and Simmons (1980) described a case in which two males accompanied a female when in- specting a nest. Here, we describe two different cases of cooperative breeding by hobbies in the North of Spain. In 2000, we begun monitoring hobbies in Bizkaia, Northern Spain, and 41 pairs were found. In 2001, two of these territories were occupied by pairs, but were oc- cupied by trios in 2002. One territory was located in a large pine (Pinus radiata) plantation in Muskiz and the other was in a large eucalyptus {Eucalyptus globulus) plan- tation in Derio. The pair at Muskiz bred in a pine situated in the mid- dle of the plantation in 2001 and they produced two fledglings. In 2002, we found the nest located only 5 m away from the previous year’s nest. On 4 August 2002, we witnessed three hobbies flying together for 5 hr at the pine plantation territory. One hobby (the female) was obviously larger than the other two (the males); these sizes differences were evident, mainly when the falcons were flying together (Chapman 1999). The female and one male were then aged as 3+ yr and the other male was a second-year bird (Forsman 1997). They, together, mobbed two Common Buzzards {Buteo huteo). Later, one of the males captured a passerine and passed it to the female, who ate it while flying near the two males. After- wards, the female and one of the males perched on the top of the trees close to the nest. Thirty minutes later, the other male approached and flew over the perched falcons. Then the two males left together, while the fe- male went to the nest. On 10 August, we observed one hobby 1 km away from the nest carrying a Common Swift {Apus apus) . This fal- con approached the nest after perching in a pine to pluck the bird. Two hours later, we saw one of the males trying to hunt another swift, then join the female and mob three different Common Buzzards. On 11 August, the three hobbies attacked an Eurasian Eagle-Owl {Bubo bubo) that was 20 m from the nest. On 31 August, two chicks were flying, but always stayed close to the nest dur- ing a 5 hr period, while the female and the two males perched near the chicks. The three adult hobbies never showed signs of aggression between themselves. In the territory at Derio in 2001, we observed a pair of hobbies, but could not 6nd the nest because of the dense tree canopy. In 2002, we spent 3 d searching for the nest. On the first 2 d (27 and 28 July) , we saw three hobbies flying together, one adult female (larger) and two males. We could not see the tail feathers of the males clearly, so we could not determine their ages (Forsman 1997). On the second day, the falcons flew together for a period of 5 hr. On the third day (1 August), under a persistent rain, we saw only one hobby and found the nest in a eucalyptus tree in the middle of the plantation. On 3 August, we climbed to the nest and found it empty, and did not see any hobbies while we were there. As we were leaving the plantation, we saw a Booted Eagle {Hi- eraaetus pennatus) flying above, and the three hobbies at- tacked it. The falcons chased the eagle for more than 1 km, and then returned to the plantation. Based on our observations, we could not distinguish whether these trios represented polyandrous breeding or helping behavior. It is possible that the second males m the Bizkaia trios could have been yearlings born in the same territories, who returned and then helped to de- fend the territory. By helping, these yearlings could gain experience that wonld prove useful in their third calen- dar year, when male hobbies usually reach sexual matu- rity. Most second-year hobbies behave as floaters (Chap- man 1999), and they are strongly philopatric, as shown by the fact that they retnrn from their wintering areas to within a mean of 10 km from their natal sites (Chapman 1999), which could lead to helping behavior. The age of the second male at the pine plantation nest was consis- tent with the interpretation that this bird was a helper. The benefits of the helping behavior for the adult pair were not clear (Donazar 1993). In the two trios moni- tored, reproductive success was relatively low (one of two nests fledged a total of two young) , but this sample was extremely small. Resumen. — En el aho 2000 comenzamos un estudio so- bre el alcotan europeo {Falco subbuteo) en Bizkaia, Norte de Espana. Localizamos 41 territorios. Un ano despues se controlaron varias parejas, a las que se realize un es- trecho seguimiento. En el aho 2002 continuamos el se- guimiento y encontramos que en dos de los territorios controlados un aho antes, habia trios poliandricos, es de- cir, dos machos y una hembra compartiendo las labores de defensa de territorio y reproduccion. Realizando un seguimiento mas exhaustivo, comprobamos que los dos machos de cada territorio participaban en la ceba y ex- pulsaban ferozmente a cualquier intruso que se acercase al nido. En una de las parejas pudimos comprobar que la hembra y uno de los machos eran individuos adultos. 344 Short Communications VoL. 37, No. 4 raientras que el otro era un ejemplar nacido el ano an- terior, por lo que sospechamos que podria ser un polio de la pareja que colabora con sus progenitores para sacar adelante la nidada y, de esta forma, va cogiendo expe- riencia mientras llega a la madurez sexual que se pro- duce un ano despues. [Traduccion de los autores] Acknowledgments We would like to thank J.C. Bednarz, J.A. Donazar, and two anonymous reviewers for providing helpful com- ments on earlier drafts of this manuscript. We acknowl- edge Lander Astorkia, Fernando Ruiz Moneo, and Javier Elorriaga for the field assistance. Literature Cited Alcock, J. 1993. Animal Behavior, 5th Ed. Sinauer Asso- ciates, Sunderland, MA U.S.A. Bijlsma, R. 1980. De Boomvalk. Kosmos, Utrecht, Ant- werpen, Belgium. Chapman, A. 1999. The Hobby. Arlequin Press, Essex, U.K. Cramp, S. and K.E.L. Simmons (Eds.). 1980. Handbook of the birds of Europe, the Middle East, and North Africa, Vol. 2. Oxford Univ. Press, Oxford, U.K. DonAzar, J.A. 1993. Los buitres ibericos. Biologia y con- servacion. J.M Reyero, Madrid, Spain. Forsman, D. 1997. The raptors of Europe and Middle East. A handbook of Field Identification. T. & A.D. Poyser, London, U.K. Kimball, R.T., P.G. Parker, and J.C. Bednarz. 2003. The occurrence and evolution of cooperative breeding among the diurnal raptors (Accipitridae and Falco- nidae). Auk 120:717-729. Korpimaki, E,, K. Lahti, C.A. May, D.T. Parkin, G.B. Powell, P. Tolonen, and J. Wetton. 1996. Copula- tory behaviour and paternity determined by DNA fin- gerprintings in kestrel: effects of cyclic food abun- dance. Anim. Behav. 51:945—955. Marks, J.S., J.L. Dickinson, and J. Haydock. 1999. Ge- netic monogamy in Long-eared Owls. Condor 101:854— 859. Rohner, C. 1997. Non-territorial “floaters” in Great Horned Owls: space use during a cyclic peak of snow- shoe hares. Anim. Behav. 53:901-912. Received 15 November 2002; accepted 13 August 2003 /. Raptor Res. 37(4):344-349 © 2003 The Raptor Research Foundation, Inc. Food Habits of Peregrine Falcons in Kentucky Kristina M. Garter,^ Michael J. Lacki, and Matthew R. Dzialak Department of Forestry, University of Kentucky, Lexington, KY 40546 U.S.A. Laura S. Burford Kentucky Department of Fish and Wildlife Resources, Wildlife Diversity Program, Frankfort, KY 40601 U.S.A. Ronald O. Bethany Louisville Gas and Electric Company, Trijnble County Station, Bedford, KY 40006 U.S.A. Key Words: Peregrine Falcon] Ealco peregrinus; food hab- its] human-made habitat, Kentucky] monitoring. Many studies on Peregrine Falcons {Falco peregrinus) in- clude observations of prey taken (White et al. 2002). As a consequence of the Peregrine Falcon’s cosmopolitan distribution and adaptability, inferences derived from food habit studies often are limited to the study area in which they were conducted (Ratcliffe 1993, Schneider and Wilden 1994, Rejt 2001, Serra et al. 2001). In the Midwestern United States, a large proportion (0.70) of ^ E-mail address: kristinacarter78@hotmail.com the restored Peregrine Falcon population occupies struc- turally similar human-made breeding locations (Tordoff et al. 2001); however, land use adjacent to these breeding locations often is variable and may be reflected in the diet of the birds. For example, in Kentucky, three pairs of Peregrine Falcons occupy human-made breeding lo- cations including bridges and power plants. Land use ad- jacent to the breeding locations varies from predomi- nantly urban at one breeding location to predominantly rural at the other two breeding locations. Food-habits data from this population would be useful in enhancing our understanding of prey use among hab- itat types within the region, monitoring potential expo- December 2003 Short Communications 345 sure to contaminants in restored populations, and devel- oping future management and conservation strategies. However, few studies have quantified Peregrine Falcon food habits in this region (Myers and Pease 1995). As part of a statewide effort to monitor Peregrine Falcons, our objective was to quantify and to compare the diet of Peregrine Falcons in Kentucky among habitats and sea- sons. Study Areas and Methods The study area is a 175-km segment of the Ohio River Valley in the Outer Bluegrass Physiographic Region of Kentucky. Breeding locations include an interstate bridge spanning the Ohio River in Jefferson County (Louisville, KY; Urban I), and smokestacks associated with power plants in Trimble and Carroll counties (Rural I and II, respectively) . The topography at these sites is nearly level to moderately sloping and dominated by upland oaks (i.e., Quercus rubra, Q. alba), hickories (Carya spp.), and yellow-poplar {Liriodendron tulipifera) on the slopes and pin oak ( Quercus palustris) , eastern cottonwood {Populus deltoides) , silver maple {Acer saccharinum) , and eastern syc- amore {Platanus occidentalis) in the floodplains. Land use adjacent to these breeding locations was the most notable difference among sites. To characterize general landscape attributes at these sites, we used a geo- graphic information system (CIS; ArcView®, ESRI, Red- lands, California) and Kentucky and Indiana GAP anal- ysis Program data (Center for Remote Sensing and GIS 1996, Mid-American Remote Sensing Center 2001) to de- termine the proportion of three land use classes, includ- ing urban/developed, agriculture, and forest, within a 10-km radius of each breeding location. We collected prey remains and pellets, and recorded visual observations of foraging Peregrine Falcons at ter- ritorial sites during March 1999-December 2001. During spring and summer. Peregrine Falcons bring much of their prey to the nest associated with courtship and the feeding of young, allowing a greater number of speci- mens to be collected compared to fall and winter (Rat- cliffe 1993). We examined pellet contents according to the methods of Sabo and Laybourne (1994). We com- pared prey remains to museum specimens for identifi- cation (Oro and Telia 1995). Each species identified in a pellet or as a prey remain was considered to be one occurrence, unless multiples of the same body part were present (Mersmann et al. 1992). We calculated and compared percent occurrence and percent biomass of prey taken over all sites, among sites and seasons, and according to seasonal status of the prey (i.e., resident, summer resident, winter resident, tran- sient; Mengel 1965). For analyses among seasons, we clas- sified samples as spring (March-May), summer (June- August), fall (September-November) , and winter (December-February) . We calculated percent biomass by multiplying the number of individuals in a species by the mean mass for that species and then divided that value by the total biomass of all species (Dunning 1984, Corser et al. 1999). We calculated Simpson’s index of diversity (1-D) and compared diversity indices among sites and seasons (Krebs 1989). Simpson’s index varies between 0 and 1, with diversity increasing as values approach one. Table 1. Percent occurrence and percent biomass of prey taken by three pairs of Peregrine Falcons in Ken- tucky based on prey remains, pellets, and observations of prey captures {N = 465), March 1999-December 2001 ® Scientific names of bird species are given in Appendix Species Frequency (%) Biomass (%) European Starling*’’' *^ 35 14 Rock Dove*’ ' '^ 27 70 Eastern Meadowlark’’ ^’*' 8 3 Blue Jay’’*^’'' 7 3 Brown-headed Cowbird' *' 5 1 Red-winged Blackbird'-'’ 4 1 Northern Flicker'’'’ 3 2 Common Grackle’’ ' '’ 2 1 Mourning Dove’’ ' *’ 2 1 Northern Mockingbird'’'’ 2 0.4 ® Species comprising <1% of observations (0. 1-0.7% of total bio- mass) included American Robin‘S, Killdeer^, Red-bellied Wood- pecker‘s, Eastern Towhee‘='‘^, House Sparrow', Wood Thrush‘S, Yel- low-billed Cuckoo'’, Northern Cardinal', Scarlet Tanager', Northern Rough-winged Swallow', American Coot', Pectoral Sandpiper*', Bonaparte’s GulU, Common Nighthawk*', Brown Thrasher*', White-eyed Vireo*', Cedar WaKwing**, Hairy Wood- pecker*', Baltimore Oriole*^, and a baU. '* Species recorded at Jefferson County, KY (Urban I). *= Species recorded at Trimble County, KY (Rural I) . *' Species recorded at Carroll County, KY (Rural II) . We derived 95% confidence intervals for Simpson’s in- dex using a jackknife resampling technique (Krebs 1989), and we report results as Simpson’s index ± 95% confidence interval. Results General landscape attributes differed among breeding locations. Urban I is predominantly urban/developed (0.70), with some agriculture (0.14), and little forest (0.07 of landscape within 10-km radius). Rural I and Ru- ral II are largely agriculture (0.48 and 0.46, respectively) and forest (0.44 and 0.47, respectively), with litde urban- ization/development (0.02 and 0.01, respectively). In all, we collected 465 samples {N = 384 prey remains, V = 54 pellets, N = 27 observed prey captures). We col- lected 212 samples at Rural I, 192 at Rural II, and 61 at Urban I. At Urban I, many prey remains were inaccessi- ble and likely fell into the Ohio River. We identified 21 different prey species at Rural I, 20 at Rural II, and six at Urban I (Table 1). We identified 20 different prey spe- cies during spring, 24 during summer, five during fall, and seven during winter. Across all sites and seasons com- bined, Rock Doves {Columba livia) and European Star- lings (Sturnus vulgaris) were the most frequent prey and comprised 84% of the total biomass in the diet of Pere- 346 Short Communications VoL. 37, No. 4 Table 2. Comparison of percent occurrence and percent biomass of prey taken among three pairs of Peregrine Falcons in Kentucky based on prey remains, pellets, and observations of prey captures {N = 465), March 1999- December 2001.^ Scientific names are given in Appendix. Species Jefferson County (Urban I; A = 61) Trimble County (Rural I; A = 212) Carroll County (Rural II; A = 192) Frequency (%) Biomass (%) Frequency (%) Biomass (%) Frequency (%) Biomass (%) European Starling 12 2 18 8 60 32 Rock Dove 75 95 23 67 16 56 Eastern Meadowlark 0 0 13 6 3 2 Blue Jay 0 0 9 4 7 4 Brown-headed Cowbird 0 0 9 2 0 0 Red-winged Blackbird 0 0 9 3 0 0 Percent occurrence and percent biomass of all other species was :^5%. grine Falcons in Kentucky (Table 1). Peregrine Falcons preyed heavily on resident birds (>97%). At Urban I, Rock Doves and European Starlings were the most frequent prey, but Rock Doves comprised 95% of the biomass in the diet (Table 2). At Rural I, Rock Doves, European Starlings, and Eastern Meadowlarks {Sturnella magna) had the highest percent occurrence and together comprised 81% of the biomass in the diet (Table 2). At Rural II, European Starlings were the most frequent prey; however. Rock Doves were a majority of the biomass in the diet of falcons (Table 2). The diet of at Rural I was most diverse {1-D = 0.88 ± 0.01), followed by Rural 2 (1-D = 0.63 ± 0.04), and Urban I (1-D = 0.52 ± 0.24). In spring (N = 112), European Starlings, Eastern Meadowlarks, and Rock Doves were the most frequent prey taken and comprised the majority of the biomass in the diet (Table 3). In summer {N = 319), European Star- lings and Rock Doves together comprised s70% occur- rence and >90% of the biomass in the diet (Table 3). During fall {N = 14), European Starlings and Blue Jays {Cyanocitta aristata) were the most frequent prey and com- prised the majority of the biomass in the diet (Table 3). In winter {N = 20), European Starlings were the majority of the diet in terms of occurrence and biomass (Table 3). The diet of Peregrine Falcons was most diverse in spring (1-D = 0.88 ± 0.01), slightly less diverse in fall and summer (1-D = 0.80 ± 0.06, and 1-D = 0.75 ± 0.01, respectively), and least diverse in winter (1-D = 0.62 ± 0.15). Discussion Results of this study are consistent with other research that identified Rock Doves as a primary prey item in the Table 3. Comparison of percent occurrence and percent biomass of prey taken by Peregrine Falcons among seasons m Kentucky based on prey remains, pellets, and observations of prey captures (by/ N = 465), March 1999-December 2001.“ Scientific names of bird species are given in Appendix. Species Spring (A = 112) Summer (A = 319) Fall (A = 14) Winter (A = 20) Freq. (%) Biomass (%) Freq. Biomass (%) (%) Freq. (%) Biomass (%) Freq. Biomass (%) (%) European Starling 22 12 37 13 36 . 19 65 40 Rock Dove 14 51 33 78 14 50 5 20 Eastern Meadowlark 18 10 0 0 0 0 5 3 Blue Jay 13 7 0 0 29 16 10 6 Common Crackle 7 5 0 0 0 0 0 0 Northern Flicker 5 5 0 0 14 12 5 5 Brown-headed Cowbird 0 0 6 1 7 2 0 0 Red-winged Blackbird 0 0 5 1 0 0 0 0 Red-bellied Woodpecker 0 0 0 0 0 0 5 2 American Coot 0 0 0 0 0 0 5 24 “^Percent occurrence and percent biomass of all other species was ^5%. December 2003 Short Communications 347 diet of Peregrine Falcons in urban and industrial habitats (Barber and Barber 1988, Myers and Pease 1995, Cade et al. 1996a). Worldwide, Peregrine Falcons consistently rely on Columbidae as their main prey (Ratcliffe 1993, White et al. 2002) . Other species recorded in the diet of urban Peregrine Falcons represented locally abundant and resident species (Barber and Barber 1983, 1988, Bell et al. 1996, Rejt 2001). Peregrine Falcons in Kentucky exhibited similar tendencies; however, the percent oc- currence and biomass of European Starlings in the diet of Peregrine Falcons at Rural I and II were higher than in other studies of urban and non-urban areas of (4.1— 11.2% occurrence and 3-5% biomass) in North America (Barber and Barber 1988, Myers and Pease 1995, Corser et al. 1999, Rejt 2001). Herbert and Herbert (1965) re- ported Peregrine Falcons feeding regularly on European Starlings when they nested near roosting colonies. We observed roosting colonies of European Starlings at both power plants in this study, which may have accounted for their frequent occurrence in the diet of these pairs. Throughout their range, Peregrine Falcons have been documented to prey on a diverse assemblage of taxa (Rat- cliffe 1993, White et al. 2002); however, few studies have compared the diversity of prey in the diets of Peregrine Falcons among territorial pairs. Based on description data, Bell et al. (1996) suggested that one pair of Pere- grine Falcons appeared to exhibit a more diverse diet than two other pairs that nested on bridges in the vicinity of San Francisco Bay, CA. Similarly, in our study Pere- grine Falcons at Rural I had a more diverse diet com- pared to the other rural pair. The proximity of Rural I and Rural II to each other and their similarities in breed- ing structure type (i.e., power plants) would imply similar availability of prey assemblages. The differences in diver- sity we observed between these two pairs may be attri- buted to a difference in preference. Several studies ex- amining the diet of Peregrine Falcons have demonstrated individual preferences for a specific prey type. For ex- ample, Cade et al. (1996a) surveyed Midwestern Pere- grine Falcon food habits and identified cases in which some birds concentrated on single species such as Com- mon Nighthawks {Chordeiles minor) in Minnesota or cuck- oos {Cuculus sp.) in Wisconsin. Nonetheless, information on prey species abundance would be necessary to deter- mine whether differences were attributed to availability of prey or to preference. Seasonal variation in the diet of Peregrine Falcons could likely represent differences in prey availability (Cade et al. 1996a, Serra et al. 2001). For example, mi- gratory species increase in frequency in the diets of Per- egrine Falcons during spring and fall, whereas the per- centage of resident birds decreases (Ratcliffe 1993, Cade et al. 1996a, Rejt 2001, Serra et al. 2001). Results of this study reflected the lowest frequency of Rock Doves and European Starlings in the spring coinciding with the presence of migratory species such as Common Night- hawks and Scarlet Tanagers (Piranga olivacea). Also, the presence of migratory species during spring coincided with an increase in diversity in the diet. Some species may be more conspicuous to Peregrine Falcons in the spring because of their mating displays. For example, the in- crease in the frequency of Eastern Meadowlarks in this study may be attributed to flashy mating displays exhib- ited in the spring. Monitoring efforts are important to endangered spe- cies recovery (Cade et al. 1996b), Continued vigilance is necessary to assure long-term success of restored popu- lations. For example, although there are no data linking use of urban or industrial breeding locations to repro- ductive maladies, potential contaminants and possible routes of exposure through prey should be examined (e.g., Fimreite et al. 1970, DeMent et al. 1986, Cade and Bird 1990, Mora et al. 2002). Monitoring food habits and other detailed aspects of feeding ecology in restored Per- egrine Falcon populations can be useful in detecting long-term population exposure to food-related threats and in developing proactive management strategies. Resumen. — Colectamos restos de presas y egagropilas de sitios nido, e hicimos observaciones visuales de captura de presas {N = 465) para tres parejas de halcon peregri- ne {Falco peregrinus) que anidaron en Kentucky de marzo a diciembre de 2001. Las aves residentes dan cuenta del 97% de la dieta del halcon, y especificamente la paloma zorita {Columba livia) y el estornino europeo {Sturnus vul- garis) abarcan el 62% de las presas tomadas. La dieta de una pareja rural del halcon peregrine fue mas diversa (indice de diversidad = 0.88 ± 0.01) que la de otras parejas. La dieta fue mas diversa en primavera (indice de diversidad = 0.88 ± 0.01) comparada con otras estacio- nes. A pesar de las diferencias en el habitat (urbane ver- sus rural), los halcones peregrines consistentemente hi- cieron presa sobre estorninos europeos y palomas zoritas mas que sobre otras especies. Los esfuerzos de monitoreo para restaurar las poblaciones de halcones peregrines pueden beneficiarse de estudios cuantitativos de los ha- bitos alimenticios, especialmente donde las especies de presa consumidas pueden ser indicadores de la calidad ambiental. [Traduccion de Cesar Marquez] Acknowledgments Funding for this project was provided by a grant from the Kentucky Department of Fish and Wildlife Resources and by the University of Kentucky College of Agriculture. We thank the Kentucky Utilities Generating Station at Ghent, KY, and Louisville Gas and Electric Company, Trimble County Station, Bedford, KY, for access to the sites and cooperation in collecting data. We thank L. Pe- rez for assistance in collection of data. T. Cade, H. Tor- doff, and D. Dekker provided insightful reviews of an ear- lier version of this manuscript. This research (KAES 02-09-55) is connected with a project of the Kentucky Agricultural Experiment Station and is published with the approval of the Director. 348 Short Communications VoL. 37, No. 4 Literature Cited Barber, J.C. and M.M, Barber. 1983. Prey of an urban Peregrine Falcon. MD BirdLife 39:108-110. AND . 1988. Prey of an urban Peregrine Falcon — part II. MD BirdLife 44:37-39. Bell, D.A., D.P. Gregoire, and B.J. Walton. 1996. Bridge use by Peregrine Falcons in the San Francisco Bay area. Pages 15-24 in D. Bird, D. Varland, and J. Negro [Eds.], Raptors in human landscapes. Academ- ic Press, San Diego, CA U.S.A. Cade, T.J. and D.M. Bird. 1990. Peregrine Falcons, Falco peregrinus, nesting in an urban environment: a review. Can. Field-Nat. 104:209-218. , M. Martell, P. Redig, G.A. Septon, and H.B. Tordoff. 1996a. Peregrine Falcons in urban North America. Pages 3-14 in D. Bird, D. Varland, and J. Negro [Eds.], Raptors in human landscapes. Academ- ic Press, San Diego, CA U.S.A. , J.H. Enderson, and j. Linthicum. 1996b. Guide to management of Peregrine Falcons at the eyrie. The Peregrine Fund, Boise, ID USA. Center for Remote Sensing and GIS. 1996. Indiana landcover/landuse. Indiana State University, Terre Haute, IN U.S.A. Corser, J.D., M. Amaral, C.J. Martin, and C.C. Rimmer. 1999. Recovery of a cliff-nesting Peregrine Faicon, Fal- co peregrinus, population in northern New York and New England, 1984—1996. Can. Field-Nat. 113:472—480. DeMent, S.H., J.J. Chisolm Jr., J.C. Barber, and J.D. Strandberg. 1986. Lead exposure in an “urban” Per- egrine Falcon and its avian prey./. Wildl. Dis. 22:238- 244. Dunning, J.B., Jr. 1984. Body Weights of 686 species of North American birds. Western Bird Banding Associ- ation Monograph No. 1. Eldon Publishing, Cave Creek, AZ U.S.A. Fimreite, N., R.W. Fyfe, and J.A. Keith. 1970. Mercury contamination of Canadian prairie seed eaters and their avian predators. Can. Field-Nat. 84:269—276. Herbert, R.A. and K.G.S. Herbert. 1965. Behavior of Peregrine Falcons in the New York City region. Auk 82:62-94. Krebs, C.J. 1989. Ecological methodology. Harper Collins Publishers, New York, NY U.S.A. Mengei., R.M. 1965. The birds of Kentucky. Ornithol. Monogr. No. 3. Allen Press Inc., Lawrence, KS U.S.A. Mersmann, T.J., D.A. Buehi.er, J.D. Fraser, and J.K.D. Seegar. 1992. Assessing bias in studies of Bald Eagle food habits./. Wildl. Manage. 56:73-78. Mid-American Remote Sensing Center. 2001. Vegetation map of Kentucky. Murray State University, Murray, KY U.S.A. Mora, M., R. Skiles, B. McKinney, M. Paredes, D. Buck- ler, D. Papoulias, and D. Kitin. 2002. Environmental contaminants in prey and tissues of the Peregrine Fal- con in the Big Bend Region, Texas, U.S.A. Environ. Pollut. 116:169-176. Myers, L.M. and J.L. Pease. 1995. Analysis of Peregrine Falcon {Falco peregrinus) food habits. Iowa BirdLife 2: 25-29. Oro, D. and j. Telia. 1995. A comparison of two meth- ods for studying the diet of the Peregrine Falcon. /. Raptor Res. 29:207-210. Ratcliffe, D. 1993. The Peregrine Falcon, 2nd Ed. T. & A. D Poyser, London, U.K. Rejt, L. 2001. Feeding activity and seasonal changes in prey composition of urban Peregrine Falcons Falco pe- regrinus. Acta Ornithol. 36:165-169. Sabo, B.A. and R.C. Laybourne. 1994. Preparation of avi- an material recovered from pellets and as prey re- mains./. Raptor Res. 28:192—193. Schneider, R. and I. Wilden. 1994. Choice of prey and feeding activity of urban Peregrine Falcons Falco pere- grinus during the breeding season. Pages 203-209 in B. -U. Meyburg and R.D. Chancellor [Eds.], Raptor conservation today. Pica Press, East Sussex, U.K. Serra, G., M. Lucentini, and S. Romano. 2001. Diet and prey selection of nonbreeding Peregrine Falcons in an urban habitat of Italy. / Raptor Res. 35:61-64. Tordoff, H.B., M.S. Martell, P.T Redig, and MJ. So- LENSKY. 2001. Midwest Peregrine Falcon restoration, 2000 report. Bell Museum of Natural History/The Raptor Center. University of Minnesota, St. Paul, MN U.S.A. White, C.M., NJ. Clum, TJ. Cade, and W.G. Hunt. 2002. Peregrine Falcon {Falco peregrinus) . In A. Poole and F. Gill [Eds.], The Birds of North America, No. 660. The Birds of North America, Inc., Philadelphia, PA U.S.A. Received 14 May 2003; accepted 24 July 2003 December 2003 Short Communications 349 Appendix. Common and scientific names of prey taken by three pairs of Peregrine Falcons nesting in Kentucky based on observations of prey remains, pellets, and prey captures (N = 465), March 1999-December 2001. Family Scientific Name Common Name Rallidae Fulica americana American Coot Charadriidae Charadrius vociferus Killdeer Scolopacidae Calidris melanotos Pectoral Sandpiper Laridae Larus Philadelphia Bonaparte’s Gull Columbidae Columba livia Rock Dove Zenaida macroura Mourning Dove Cuculidae Coccyzus americanus Yellow-billed Cuckoo Caprimulgidae Chordeiles minor Common Nighthawk Picidae Picoides villosus Hairy Woodpecker Melanerpes carolinus Red-bellied Woodpecker Colaptes auratus Northern Flicker Vireonidae Vireo gtiseus White-eyed vireo Corvidae Cyanocitta cristata Blue Jay Hirundinidae Stelgidopteryx serripennis Northern Rough-winged Swallow Turdidae Hylocichla mustelina Wood Thrush Turdus migratorius American Robin Mimidae Mimus polyglottos Northern Mockingbird Toxostoma rufum Brown Thrasher Sturnidae Sturnus vulgaris European Starling Bombycillidae Bombycilla cedrorum Cedar Waxwing Thraupidae Piranga olivacea Scarlet Tanager Emberizidae Pipilo erythrophthalmus Rufous-sided Towhee Cardinalidae Cardinalis cardinalis Northern Cardinal Icteridae Sturnella magna Eastern Meadowlark Agelaius phoeniceus Red-winged Blackbird Quiscalus quiscula Common Crackle Molothrus ater Brown-headed Cowbird Icterus galbula Baltimore Oriole Passeridae Passer domesticus House Sparrow Vespertilionidae Myotis sp. Bat Letter J Raptor Res. 37(4):350-351 © 2003 The Raptor Research Foundation, Inc. Harris’s Hawk Dies After Colliding with Motor Vehicle The extent to which raptors are killed on roads may depend on a variety of biotic and abiotic factors (e.g., hunting habits, dispersal behavior, vehicular traffic, motor speed) . Vehicle mortality is a main source of mortality in popula- tions of Barn Owl, Tyto alba, (Newton et al. 1999, Mortality causes in British Barn Owls {Tyto alba), based on 1101 carcasses examined during 1963-1996. Pages 299-307 mJ.R. Duncan, D.H. Johnson, and T.H. Nicholls [Eds.] , Biology and Conservation of Owls of the Northern Hemisphere, Second International Symposium, U.S. Department of Ag- riculture, St. Paul, MN U.S.A.) and Red-tailed Hawk, Buteo jamaicensis (Reran 1981, Raptor Res. 15:108-112). Motor collision deaths have also been reported in large, diurnal raptors such as the Ferruginous Hawk, B. regalis, (Bechard and Schmutz 1995, In A. Poole and F. Gill, [Eds.], The birds of North America, No. 172. The Academy of Natural Sciences, Philadelphia, PA and the American Ornithologists’ Union, Washington, DC U.S.A.), Red-shouldered Hawk, B. lineatus, (Reran 1981), and Broad-winged Hawk, B. platypterus (Loos and Kerlinger 1993,/. Raptor Res. 27:210-213). Nocturnal hunting habits may increase some species’ risk of colliding with motor vehicles. For example, bright headlights may temporarily blind owls hunting at night. In support of this idea, the frequency of road mortality in the Little Owl {Athene noctua) was slightly lower (2.3%) along illuminated roads (Hernandez 1988,/. Raptor Res. 22: 81—84). Species that habitually scavenge road-killed animals are more likely to die due to collisions (Brown and Amadon 1968, Eagles, hawks, and falcons of the world, Vol. I. McGraw-Hill Book Co., New York, NYU.S.A.). Within species, certain age and gender classes may be more prone to this type of mortality; specifically, young birds under- going post-fledging dispersal may be more susceptible to vehicular collisions (Hernandez 1988, Loos and Kerlinger 1993, Massemin et al. 1998, Ibis 140:70-75). Poor body condition may increase the likelihood of being struck by a vehicle. However, the body condition of road-killed carcasses compared to captive immature Barn Owls fed ad libitum differed only for mature females (Massemin et al. 1998). The type of roadside vegetative cover and available perch height may also contribute to the likelihood that a raptor may be killed on the road (Hernandez 1988) . Newton et al. (1999) documented an increase in the proportion of mortality due to collision with vehicles over time in the Barn Owl. They attributed this increase to the construction of more roads, increased traffic volume, and higher vehicle speeds. The extent to which the susceptible raptor species are killed by vehicles may warrant further study where highway infrastructure is expanding. Quantitative data on the causes for mortality in the Harris’s Hawk {Parabuteo unicinctus) are lacking (Bednarz 1995, In A. Poole and F. Gill [Eds.], The birds of North America, No. 146. The Academy of Natural Sciences, Philadelphia, PA, and the American Ornithologists’ Union, Washington, DC U.S.A.). Documented cases of mortality in fledged and adult Harris’s Hawks include deaths by shooting. Great Horned Owl {Bubo virginianus) predation, illegal trapping (Bednarz 1995), electrocution, probable drowning in livestock water tanks, and injuries sustained from cholla cactus {Opuntia sp.) that probably would have been fatal had not the researcher intervened (Whaley 1986, Raptor Res. 20: 1-15). We provide the first report of a Harris’s Hawk dying from collision with a motor vehicle. On 22 November 2002, we found a dead adult female Harris’s Hawk on the ground, 10 m from U.S. Highway 287, Electra, Wichita County, Texas. The hawk had been dead for ca. 3 d. The body lay prostrate, with the head facing away from the road and the tail pointing upward, emerging above the grass, and directed toward the road. The bird was at least 3 yr old judging from its plumage of only adult feathers from two different years (some rectrices and remiges were extremely worn and faded while others were bright and intact) . We examined the specimen and found a compound fracture at the proximal end of the left humerus; the humerus was broken in two pieces. The bird appeared to have been in very fit condition. Pectoral muscles protruded on either side of the keel, and a layer of subcutaneous fat covered much of the abdomen. The location of the bird and the nature of its injury implied that It died after colliding with a motor vehicle; there was no other structure nearby with which the bird could have collided. Also of interest is the location of this specimen outside the reported breeding range. The nearest published population of Harris’s Hawks is ca. 290 km away (Bednarz 1995). We did not search the area for additional Harris’s Hawks. The habitat consisted of 3-4 m high mesquite {Prosopis glandulosa) and appeared to be suitable for the species (Bednarz 1995). The frequency of Harris’s Hawk deaths from collisions with motor vehicles is probably very low (J. Bednarz and 350 December 2003 Letter 351 W. Whaley pers. comm.). Harris’s Hawks rarely feed on road-killed animals. During his 10-yr study in southeastern New Mexico, J. Bednarz observed only two instances of scavenging, and both involved immature birds in late winter or early spring. The Harris’s Hawk’s apparent avoidance of road killed animals probably reduces its risk of motor vehicle collision (J. Bednarz pers. comm.). Nonetheless, Harris’s Hawks often perch on utility poles and wires, fence posts, and signs along roads. Typical quarry are ground-dwelling mammals and generally low-flying birds associated with scrub (Bednarz 1995). Thus, collision with motor vehicles seems a likely source of mortality for populations of Harris’s Hawks living around humans. Moreover, we predict that the frequency of such deaths will increase as Harris’s Hawk habitats become more developed by humans. We thank J. Bednarz, M. Patten, W. Whaley, and H. Woodward for helpful suggestions that improved this manu- script. — -Jennifer O. Coulson, Department of Ecology and Evolutionary Biology, Tulane University, 310 Dinwiddie Hall, New Orleans, LA 70118 U.SA.; e-mail address: jacoulson@aol.com; and Thomas D. Coulson, Department of Biochemistry, Louisiana State University Medical Center, 1100 Florida Avenue, New Orleans, LA 70119 U.S.A. BOOK REVIEW J Raptor Res. 37(4):352-353 © 2003 The Raptor Research Foundation, Inc. North American Owls: Biology and Natural His- tory, Second Edition. By Paul A. Johnsgard. 2002. Smithsonian Institution Press, Washington, DC. xiii + 298 pp., 42 color plates, 12 tables, 76 text fig- ures, 3 appendices. ISBN 1-56098-939-4. Cloth, $49.95. — The first edition of this book, published in 1988, was widely acclaimed as one of the finest books on North American owls. I described it as a well-balanced coverage of the biology and natural history of the North American species {Wilson Bull. 102:192-193, 1990). Now, 14 years later, we see a revised edition, and I will focus primarily on how the new book differs from the first edition. The new edition is essentially the same size as the first and only three pages longer despite add- ing coverage of 12 Mexican species, 24 new figures, 12 new color plates, about 15% more text, and some Internet sources on owls. Packing all this new information in without increasing the size of the book was accomplished by eliminating detailed plumage descriptions and making some changes in the format; double-column format versus single column, smaller font size, and reducing the size of figures and tables. Johnsgard also nearly doubled the number of references, but admits that some of the new references are not cited in the text, a mis- take in my opinion because important new infor- mation has not been integrated into the book. Organization of the second edition, like the first, consists of two parts. Part one is a general review of owl evolution and systematics, ecology, mor- phology, physiology, behavior, reproduction, and folklore, and part two contains detailed natural his- tories of the 31 species. How well did Johnsgard pull off this revision? Unevenly, I think, is the best word to describe it. On the one hand, large sections in the first part are unaltered other than some minor editing. For example, the sections on vocalizations and mythol- ogy are essentially unchanged. On the other hand, the section on morphological characteristics has several new figures and a whole new paragraph on leg and foot musculature, but its source is not cit- ed. I am left wondering if the asymmetrical updat- ing was based on the relative amounts of new ma- terial in the literature or some other criterion. The same imbalance of revision applies to the species accounts. The Spotted Owl {Strix occidental- is) account contains much new material, reflecting the intense research on it since the first edition came out. The Eastern Screech-Owl {Otus asio) ac- count is also heavily updated with information from many new sources, but the Great Horned Owl {Bubo virginianus) account is not. Only a few new sources are cited adding very little new informa- tion. The Barn Owl {Tyto alba), the species I am most familiar with, was updated primarily from Tay- lor’s Barn Owls: Predator-Prey Relations and Conser- vation (Cambridge Univ. Press, Cambridge, UK, 1994); although an excellent book, it is concerned mostly with European birds, I was disappointed to find that several of my recent publications on Barn Owl reproduction, mate choice, and morphology {Auk 107:246-254, 1990; Condor 96:468-484, 1994; Auk 114:581-592, 1997) were not even listed in the reference section; others were listed but not cited in text. The selection process for the color images re- mains as elusive as it was in the first edition. Of the 31 species covered in the book, 26 are represented by either a painting or color photograph (all good to excellent in quality) or both, but eight species are represented by both a painting and a photo- graph. Five species, all Mexican, are not illustrated in color, and, unfortunately, the Barn, Snowy {Nyc- tea scandiaca), and Burrowing {Athene cuniculario) owls are shown only in Fuertes paintings. In my opinion, Fuertes paintings are not the best to rep- resent a species in a scientific book. Finally, why did Johnsgard use three photos of a Western Screech-Owl {Otus kennicottii) and two each of six other species? We have come to expect a lot from Johnsgard’s books. Unfortunately, this one does not quite live up to those expectations. Although it is still one of the best sources of information on North Ameri- can owls, especially with the addition of the Mexi- can species, it could have been so much more. Did Johnsgard, perhaps, anticipate this view? Why else did he pen the second sentence of the Preface: 352 December 2003 Book Review 353 “Raptoriphiles, and especially strigiphiles, are so attached to their subjects that few books escape the wrath of reviewers who feel that their particular species or subject area has been slighted or badly dealt with.” Who should buy this book? Libraries need a copy and those individuals interested in owls, who do not own a copy of the first edition, will surely want one. People who have the first edition have a harder problem — does the second edition have enough new information? The answer to that will have to be a personal one. — Carl D. Marti, Raptor Research Center, Boise State University, Boise, ID 83725 U.S.A, J. Raptor Res. 37{4):354-357 © 2003 The Raptor Research Foundation, Inc. Journal of Raptor Research INFORMATION FOR CONTRIBUTORS The Journal of Raptor Research (JRR) publishes original research reports and review articles about the biology of diurnal and nocturnal birds of prey. All submissions must be in English, but contribu- tions from anywhere in the world are welcome. Manuscripts are considered with the understand- ing that they have not been published, submitted or accepted for publication elsewhere. Manuscripts are subjected to peer review for evaluation of their significance and soundness, and edited to improve communication between authors and readers. De- cisions of the editor are final. Material is published as feature articles, short communications (usually not longer than four printed pages) , and letters (see recent issue of the JRR for examples) . Submissions that adhere closely to the JRR’s format greatly enhance the efficiency and cost of the editorial and publishing processes. Author’s efforts in this regard are deeply appreci- ated by the editorial staff. When submitting scholarly papers, send the orig- inal and three copies, a completed checklist (see below) , and a cover letter that includes: ( 1 ) a state- ment that the data in the manuscript have not been published or accepted for publication in the same form, and have not been submitted simulta- neously elsewhere, (2) the name and address of the corresponding author (in multiauthored pa- pers) including any temporary addresses where that author will be during the review process (also the phone number and, if possible, a FAX number and e-mail address of the corresponding author), and (3) if applicable, any special instructions. Au- thors may also suggest potential reviewers. If the manuscript submitted was produced on a word processor, also send a diskette (3 1/2") con- taining a single file that is identical with the print- ed copy. The electronic copy should be supplied as an IBM-compatible text file (ASCII, Word, or WordPerfect). Include information on the type of computer and word processor used. Optional elec- tronic submissions are encouraged (see General Instructions below). Manuscripts that are accepted upon condition of revision must be returned to the editor within 60 days. Manuscripts held longer will lose their pri- ority and may be treated as new submissions. The editor should be notified if extenuating circum- stances prevent a timely return of the manuscript. Authors will receive proofs of their articles prior to publication. Proofs must be read carefully to cor- rect any printer errors and returned by the fastest mail within two days of receipt TO THE EDITOR. Changes in typeset text are expensive and authors making changes, not due to printer error, will be billed for the costs. A reprint order will accompany page proofs to enable authors to buy reprints. Costs of reprints are the author’s responsibility and payment for reprints ordered must accompany the order form. Both must be sent TO THE EDITOR. Publication is expensive and member dues do not cover the entire cost of producing the JRR. Hence, the Raptor Research Foundation, Inc. ex- pects that authors defray the high costs of publi- cation through payment of page costs (currently $115.00 U.S. per page). Authors who are not as- sociated with a research institution or simply do not have access to such grants may request the page charges be waived. Such a request can only be approved if the author is a member of the Foun- dation and the article is short. Payments of amounts less than the full page charges will be ac- cepted. Authors of long manuscripts are expected to pay publishing costs. It is unlikely that articles longer than 10 printed pages or 18 typewritten pages including tables and illustrations can be pub- lished without full payment. Authors employed by government agencies, universities, or firms that will meet reprint and page charges may forward a statement to the editor indicating intent to pay. Upon receipt of such a statement, reprints will be mailed to the author and the agency will be billed with the understanding that payment will be made within 30 days. All checks should be made payable to the Raptor Research Foundation, Inc. All per- sonal payments toward publication costs are tax- deductible in the United States. 354 December 2003 Information for Contributors 355 Journal of Raptor Research CHECKLIST FOR PREPARATION OF MANUSCRIPTS {check items and submit rvith manuscript) I. General Instructions (Consult recent issues for additional guidance on format) I I Type manuscripts on one side of either 216 X 278 mm (8.5 X II") or standard international size (210 X 297 mm) good quality paper (do not use erasable or lightweight paper) . Word-processor-generated manuscripts must be done with a letter-quality or near- letter-quality printer. DOUBLE SPACE THROUGH- OUT including title, text, tables, figure legends, and literature cited. D Electronic submission. Submit e-mail message with two attached file copies of manuscript; (1) pdf file (must include figures) and (2) word processing file (Word Perfect or MS Word files accepted). Follow same format guidelines as for standard mail submis- sion. Letter of transmittal should be included in the e-mail message. E-mail to: jrr@astate.edu n Give the scientific name at the first mention of a spe- cies, both in the abstract and in the article. Scientific names of birds should follow the usage of the AOU Check-list of North American Birds (7th. ed. 1998 and subsequent supplements in the Auk) or an authorita- tive source corresponding to other geographic re- gions. Do not give subspecific identification unless it is pertinent. Capitalize first letter of words in com- plete common names for birds. Use lower case for all other common names. □ Use American spelling and Webster’s Tenth New Colle- giate Dictionary (1996, Merriam-Webster, Inc.) as a spelling authority. □ Leave at least a 25 mm (1") margin on all sides. Avoid hyphens or dashes at ends of lines; do not divide a word at the end of a line. □ Use a nonproportional font of at least elite size (4.7 characters/cm = 12 characters/inch) or 12 point, preferably Courier. DO NOT USE RIGHT JUSTIFI- CATION-LEAVE RIGHT MARGIN RAGGED. □ Use italic type for addresses, scientific names, journal names, and third level headings. □ Type last name(s) of author (s) and page number in upper right-hand corner of page 2 and all following pages. □ Cite each figure and table in the text. Do not repeat material in two forms (i.e., in text and table, or table and figure). Organize text, as far as possible, so that tables and figures are cited in numerical order. □ Use “Figure” only to start a sentence; otherwise “Fig.” if singular, “Figs.” if plural (e.g.. Fig. 1; Figs. 2, 3; Figs. 4-6). □ Use metric units throughout. □ Use these abbreviations without spelling out: hr, rain, sec, yr, mo, wk, d, km, cm, mm; designate temperature as 32°C. □ Use “continental” dating (e.g., 10 July 1993, 1-3 June, 1 1 May to 1 1 June) . □ Use 24-hour clock (e.g., 0800 H, 1345-1400 H) □ Write out numbers one to nine unless a measurement (e.g., four birds, 3 km, 40 sites, 6 yr). Use 1000 and 10 000; 0.15 instead of .15; % instead of percent. □ Each reference cited in text must be listed in the Lit- erature Cited section, and vice versa. Double check the accuracy of all entries — THE EDITORIAL STAFF CANNOT DO THIS FOR YOU. □ Literature citations in the text are as follows: a. One author-Jones (1993) or (Jones 1993) b. Two authors-Smith and Jones (1991) or (Smith and Jones 1991) c. Three or more authors-Hernandez et al. (1990) or (Hernandez et al. 1990) d. Manuscripts accepted for publication but not yet published-Howard (in press) or (Howard in press) e. Unpublished materials-K. Jacobson (unpubl data); (K. Jacobson pers. comm.); or K. Jacobson (pers. comm.) -do not place in the Literature Cited section. f. When citing several references within parentheses, separate with commas and put in chronological or- der, oldest first) . g. For manuscripts submitted as letters, place cita- tions in text in abbreviated form, e.g., (I.C. Birds 1993, J. Raptor Res. 27:45-50) . □ Assemble manuscripts for regular articles in this or- der: (1) title page, (2) abstract page, (3) text, (4) ta- bles, (5) figure legends, (6) figures. DO NOT STA- PLE. II. Title Page D Place full title 6—8 lines below top of page in all capital letters. Below title, center author’s name(s) in all capital letters and address (es) followed by a running title (short title) not to exceed 30 characters. If the author (s) is/ are currently at another location from where the work was done, use superscript number (s) fol- lowing author (s) name(s) to indicate current address in footnote at bottom of the page. In multiauthored papers, indicate the author re- sponsible for correspondence and requests for reprints. Give phone number and, if possible, FAX number and e-mail address of the corre- sponding author. III. Abstract/summary □ For regular articles, include an abstract of about 250 words in one paragraph that is completely 356 Information for Contributors VoL. 37, No. 4 without reference to the text. Be concise, in- clude the paper’s purpose, but emphasize the results. Statements like “results will be dis- cussed” are not appropriate. The abstract will also be published in Spanish. Authors fluent in both languages are encouraged to include both versions, otherwise the JRR will provide the Spanish translation. □ Include five to seven key words for indexing af- ter the abstract. □ Short communications will be printed with a Spanish summary only. Authors must provide an English summary to be translated into Span- ish unless they are fluent in Spanish. □ Avoid citing references in the abstract. If they must be cited, include journal name, volume, pages, and year, all in parentheses. IV. Text □ Follow instructions in section I. D Main headings are all capital letters and flush with left margin. □ Typical main headings for regular articles are: METHODS, RESULTS, and DISCUSSION. An introduction begins the text but does not have a heading. □ Put second-level headings in bold. Use normal indentation and capitalize first letter of each word in the second-level headline except prep- ositions and articles. □ Put third-level headings in italics. Capitalize first letter of first word only. □ Short communications and letters may or may not have headings within the text depending upon the need. V. Literature Cited □ Type references in large and small capital let- ters, including all authors’ names (e.g., Ne- meth, N.M. AND J.L. Morrison. 2002. Natal dis- persal of the Crested Caracara . . . etc. . . .). □ Verify all entries against original sources includ- ing diacritical marks and spelling in languages other than English. Capitalize all nouns in Ger- man. □ Cite references in alphabetical order by first au- thor’s surname. References by a single author precede multiauthored works by the same se- nior author regardless of date. □ List works by the same author (s) chronologi- cally, beginning with the oldest. □ Use six hyphens when the author is the same as in the preceding citation. □ “In press” citations must have been accepted for publication and must include date, volume number, and the name of the journal or pub- lisher. □ Initials of second, third, and. . . authors precede their surname. □ Abbreviate journal names according to the Se- rial Sources for the BIOSIS Data Base (published annually by the BioSciences Information Ser- vice) . □ Do not list personal communications and un- published reports. VI. Tables (Tables are expensive to print-keep them to a min- imum and put each on a separate page-try to de- sign them to fit a single column.) □ Double space throughout. Assign each table an Arabic number followed by a period. □ Table titles must be complete sentences. □ Use same size of type as in text. □ Indicate footnotes by lowercase superscript let- ters. □ Do not use vertical lines. VII. Figure Legends □ Print all figure legends on one page, double spaced. □ Number using Arabic numbers consecutively in the same order the figures appear in the text (i.e.. Figure 1., Figure 2., etc.). VIII. Preparation of Illustrations (Illustrations are referred to as figures and include drawings, graphs, and black and white half-tones [photographs] . CONSULT THE EDITOR IN AD- VANCE ABOUT COLOR.) □ Use professional standards in preparing figures; their reproduction in the JRR is virtually iden- tical to what is submitted. Consult issues of JRR for examples and Steps Toward Better Scientific Il- lustrations (Allen Press, P.O. Box 368, Lawrence, KS 66044) for more information. □ Plan figures to fit proportions in the JRR, pref- erably for a single column-printed size is 72 mm for single column width, 148 mm for full page width or 195 mm for lengthwise figures. Figures should be submitted no smaller than the final size nor larger than twice the final size (on paper no larger than 216 X 278 mm (8.5 December 2003 Information for Contributors 357 X 11") or standard international (210 X 297 mm). □ Submit drawings as original art (undiluted In- dia ink on heavy-weight, smooth-finish drafting paper) or as photomechanical transfers (PMTs) . Submit graphs as mechanical drawings or as high-resolution laser prints. Typed or handwritten text or symbols are not acceptable. Add text and symbols with press-on symbols and letters or mechanical lettering. Review copies of figures can be photographic copies but must ap- proach the quality of the original. □ Figure text must be a plain (sans serif) typeface (e.g., Helvetica), not compressed, and large enough so that it will be as large as the text type (8-10 point) when in print. □ Photographs must be sharp, high-contrast, glossy prints approximately the size that they will appear in print. If several photographs are to be included in one figure, group them butt- ed together with no space between. □ Use the same style of lettering and presentation for all figures. IX. What to Send □ Cover letter. □ Copy of this checklist completed. □ Original and three copies of manuscript and il- lustrations. □ Diskette containing a text file of the manuscript text, tables, and figures (if the manuscript was prepared using a word processor) . □ Submit to: James C. Bednarz, Editor Journal of Raptor Research Department of Biological Sciences P.O. Box 599, 117 Caraway Road Arkansas State University State University, AR 72467 U.S.A. More information? Telephone: (870) 972-3082 FAX: (870) 972-2638 E-mail: jrr@astate.edu J. Raptor Res. 37(4):358-363 © 2003 The Raptor Research Foundation, Inc, Index to Volume 37 byJaimie Varner The index includes references to general topics, species, common names, key words, and authors. Reference is also made to book reviews and letters. Taxa other than raptors are included where referenced by authors. A Accipiter brevipes, 31-36, 64—67 cooperii, 246-252 gentilis, 299—306, 337-340 gentilis laingi, 198—208 Aegolius funereus, 55-62 Aerial telemetry, 147-151 Age, 31—36 structure, 252-256 Agroecosystems, 125-134 Asio otus, 256—258 Alonso, Raul, see Zuberogoitia, Inigo Alps, Dinaric, 55-62 Anas platyrhynchos, 161-163 Anderson, Stanley H., see Smith, Roger N. Antilles, 71-75 Aquaculture facilities, 292-298 Aquila chrysaetos, 8-18 nipalensis, 75-77 Argentina, 125-134 central, 236-243 Arizona, Altar Valley, 151-157 Asia, 256-258 Asio otus, 256—258 Athene noctua, 256—258 Azkona, Ainara, see Zuberogoitia, Inigo B Balcerzak, Melissa J. and Petra Bohall Wood, Red-shoul- dered Hawk {Buteo lineatus) abundance and habitat in a reclaimed mine landscape, 188—197 Banasch, Ursula, see Rowell, Petra Barker, Michael E., see de Solla, Shane R. Bats, 244-246 Bechard, Marc J., see Canavelli, Sonia B. Bechard, Marc J. and Cesar Marquez-Reyes, Mortality of wintering Ospreys and other birds at aquaculture fa- cilities in Columbia, 292-298 Bechard, Marc J., see Yosef, Reuven Behavior, 177-187 Beingolea, Oscar, and Clayton M. White, First breeding record for Falco peregrinus in urban Lima, Peru, with remarks on the Peruvian breeding population, 84— 85 Bertran, Joan, see Margalida, Antoni Bethany, Ronald O., see Carter, Kristina M. Bibles, Brent D,, see Boal, Clint Bildstein, Keith L., see Santana, Freddy Rodriguez Bildstein, Keith, see Yosef, Reuven Blood chemistry, 299-306 Boal, Clint W., Helen A. Snyder, Brent D. Bibles, and Tracy S. Estabrook, Temporal and spatial stability of Red-tailed Hawk territories in the Luquillo Experi- mental Forest, Puerto Rico, 277-285 Boavista, 67-70 Body condition, 299-306 Bollinger, Trent, see Wayland, Mark Botella, Francisco, see Navarro, Joan Boudet, Jennifer, see Margalida, Antoni Brazil, Joe, see Dominguez, Laura Brazilian rainforest, 19-30 Breeding, 44—54 season, 236-243 Brisbin, I. Lehr, Jr., see DeVault, Travis L. British Columbia coast, 91-97 Broadcast surveys, 198-208 Brubaker, Don L., Kathleen L. Brubaker, and Bruce C. Thompson, Raptor and Chihuahuan Raven nesting on decommissioned telephone-line poles in the northern Chihuahuan Desert, 135-146 Brubaker, Kathleen L., see Brubaker, Don L. Bubo bubo, 256-258 Burford, Laura S., see Carter, Kristina M. Burgess, Neil M., see Dominguez, Laura Butastur indicus, 228-235 Buteo albicaudatus, 63—64 jamaicensis, 219—227, 277—285, 286—291 lineatus, 71-75, 177-187, 188-197 swainsoni, 125—134, 135—146 Buzzard, Grey-faced, 228-235 C Cain, Steven L., see Smith, Roger N. Calidris alpina, 91—97 California, 323-329 Call, alarm 198-208 juvenile-begging, 198-208 male-food-delivery, 198-208 Canada, 98-116 Canavelli, Sonia B., Marc J. Bechard, Brian Woodbridge, Michael N. Kochert, Juan J. Maceda, and Maria E, Zaccagnini, Habitat use by Swainson’s Hawks on 358 December 2003 Index to Volume 37 359 their austral wintering grounds in Argentina, 125- 134 Canut, Jordi, see Margalida, Antoni Capture, 44-54 Carter, Kristina M., Michael J. Lacki, Matthew R. Dzialak, Laura S. Burford, and Ronald O. Bethany, Food hab- its of Peregrine Falcons in Kentucky, 344-349 Carrete, Martina, see Navarro, Joan Cascade Mountains, 265-276 Castillo, Ihaki, see Zuberogoitia, Inigo Ceballos, Olga, see Navarro, Joan Chapman, Phillip L., see McClaren, Erica L. Chihuahuan Desert, 135-146 Raven, 135-146 Circus cyaneus, 286-291 Cliff-nesting, 323-329 Colonization, 167-172 Colton, Albie M., see Golet, Gregory H. Condon, Anne M. see Sullivan, Brian L. Conservation, 67-70 Conspecihc interaction, 333-337 Contaminants, 209-218 Cooper, Douglass M., see Sullivan, Brian L. Cooperative behavior, 342-344 Copulation behavior, 1-7 Corvus cryptoleucus, 135-146 Coulson, Jennifer O. and Thomas D. Coulson, Harris’s Hawk dies after colliding with motor vehicle, 350- 351 Coulson, Thomas D., see Coulson, Jennifer O. Court, Gordon, see Dekker, Dick Cropland, 286-291 Cuba, 71-75, 330-333 D Daniel, F. Bernard, see Dykstra, Cheryl R. Dekker, Dick, Peregrine Falcon predation on Dunlins and ducks and kleptoparasitic interference from Bald Eagles wintering at Boundary Bay, British Co- lumbia, 91—97 Dekker, Dick and Gordon Court, Gyrfalcon predation on Mallards and the interaction of Bald Eagles winter- ing in central Alberta, 161-163 Deng, Wen-Hong, Gao Wei, and Zheng Guang-Mei, Nest and roost habitat characteristics of the Grey-faced Buzzard in northeastern China, 228-235 Density, 55-62 de Solla, Shane R., Pamela A. Martin, Peter J. Ewins, and Michael E. Barker, Productivity and population trends of Ospreys in the Kawartha Lakes region, On- tario, 1978-2001, 307-314 DeVault, Travis L., Warren L. Stephens, Bradley D. Rein- hart, Olin E. Rhodes, Jr., and I. Lehr Brisbin, Jr., Aerial telemetry accuracy in a forested landscape, 147-151 Development, 37-43 Dickerman, Robert W., Talon-locking in the Red-tailed Hawk, 176 Diet, 67-70, 164-166, 177-187, 209-218, 244-246, 256- 258 Dios, Ignacio S. Garcia, Siblicide and cannibalism in the Booted Eagle {Hieraaetus pennatus) in the Tietar Val- ley, central Spain, 261 Distribution, 67-70 altitudinal, 55-62 Dominguez, Laura, William A. Montevecchi, Neil M. Bur- gess, Joe Brazil, and Keith A. Hobson, Reproductive success, environmental contaminants, and trophic status of nesting Bald Eagles in eastern Newfound- land, Canada, 209-218 Donazar, Jose Antonio, see Navarro, Joan Ducks, 91-97 sea, 164—166 Dudajek, Valerie, see Smallwood, John A. Dufty, Alfred M., Jr., see Hanauska-Brown, Lauri A. Dunk, Jeffrey R., see Smith, Roger N. Dunlin, 91-97 Dykstra, Cheryl R., Jeffrey L. Hays, Melinda M. Simon, and F. Bernard Daniel, Behavior and prey of nesting Red-shouldered Hawks in southwestern Ohio, 177- 187 Dzialak, Matthew R., see Carter, Kristina M. E Eagle, Bald, 8-18, 78-83, 91-97, 161-163, 209-218 Golden, 8-18 Steppe, 75-77 Eagle-owl, Eurasian, 256-258 Elanus caeruleus, 1—7 Elliott, John E., see Wayland, Mark Ellis, David H., Subadult and pale Steppe Eagles breed- ing in Mongolia, 75-77 Endangered species, 151-157 Estabrook, Tracy S., see Boal, Clint Ewins, Peter J., see de Solla, Shane R. Extinctions, 71-75 F Falco femoralis, 71-75 peregrinus, 91—97, 98—116, 344—349 peregiinus japonensis, 340—342 rusticolus, 161—163 sparverius, 37-43, 236-243, 286-291, 323-329 subbuteo, 342—344 Falcon, Aplomado, 71-75 Peregrine, 91-97, 98-116, 340-342, 344-349 Falconiformes, 330-333 Feeding, 340-342 Ferrero, Juan J., Juan M. Grande, and Juan J. Negro, Cop- ulation behavior of a potentially double-brooded bird of prey, the Black-winged Kite {Elanus caeruleus ) , 1-7 360 Index to Volume 37 VoL. 37, No. 4 Ferruginous Pygmy-owl, 151-157 Fetz, Trevor W., Stewart W. Janes, and Heidi Lauchstedt, Habitat characteristics of Great Gray Owl sites in the Siskiyou Mountains of southwestern Oregon, 315- 322 Finn, Sean R, see Sullivan, Brian L. Flann, Ian, Gyrfalcon color variation, 173-174 Flesch, Aaron D., Perch-site selection and spatial use by Cactus Ferruginous Pygmy-Owls in south-central Ar- izona, 151—157 Flocks, 63-64 Food habits, 236-243, 315-322, 344-349 -niche breadth, 337-340 Foraging success, 161-163 Forsman, Eric D., see Giese, Alan R. Fornasari, Lorenzo, see Yosef, Reuven Fossil birds, 71-75 Fragmentation, 117-124 Froese, Jennifer M.W., see Wayland, Mark G Galmes, Maximiliano Adrian, see Sarasola, Jose Hernan Garcia, Diego, see Margalida, Antoni Garcelon, David K., see Sullivan, Brian L. Gavrilov, Andrei, see Navarro, Joan Geographic Information System (GIS), 135-146, 147-151 Giese, Alan R. and Eric D. Forsman, Breeding season habitat use and ecology of male Northern Pygmy- Owls, 117-124 Gessaman, James A., see Wilson, Robert Gilchrist, H. Grant, see Robertson, Gregory J. Gilchrist, Sivajini, see Smallwood, John A. Glaucidium brasilianum, 151—157 cactorum, 151—157 gnoma, 117—124 Golet, Gregory H., Henry T. Golet, and Albie M. Colton, Immature Northern Goshawk captures, kills, and feeds on adult-sized Wild Turkey, 337-340 Golet, Henry T, see Golet, Gregory H. Goshawk, Northern, 198-208, 299-306, 337-340 Grand Teton National Park, 219-227 Grande, Juan M., see Ferrero, Juan J. Grant, Todd A., see Nenneman, Melvin P. Great Plains, 246-252 Guang-Mei, Zheng, see Deng, Wen-Hong Gypaetus barbatus, 157-160, 333-337 Gyrfalcon, 161-163 H Heterophils/lymphocytes ratio, 299-306 Habitat selection, 151-157, 315-322 steppe, 256-258 use, 117-124, 125-134, 188-197, 219-227, 228-235 Haliaeetus leucocephalus, 8-18, 78-83, 91-97, 161-163, 209-218 Hanauska-Brown Lauri A., Alfred M. Dufty, Jr., and Gary J. Roloff, Blood chemistry, cytology, and body con- dition in the adult Northern Goshawks (Accipiter gen- tilis), 299—306 Harrier, Northern, 286-291 Hartley, Ronald R., and Peter J. Mundy, Morphometries and status of Ayres’s Hawk-Eagle in Zimbabwe, 44- 54 Hawk, Cooper’s, 246-252 Red-shouldered, 71-75, 177-187, 188-197 Red-tailed, 219-227, 277-285, 286-291 Swainson’s, 125-134, 135-146 White-tailed, 63-64 Hawk-Eagle, Ayres’s, 44-54 Hays, Jeffrey L., see Dykstra, Cheryl R. Helper, 340-342 Hernandez, Luis Melian, see Santana, Freddy Rodriguez Hidalgo, Sonia, see Zuberogoitia, Inigo Hieraaetus ayresii, 44-54 Hilpp, G. Keith, see Moser, Brian Hiraldo, Fernando, see Navarro, Joan Hobby, Eurasian, 342-344 Hobson, Keith A, see Dominguez, Laura Holroyd, Geoffrey L,, see Rowell, Petra Home range, 219-227 Human-made habitat, 344—349 Hunting habits, 91-97 I Iraeta, Agurtzane, see Zuberogoitia, Inigo Indies, West, 71-75 Interspecific interactions, 157-160 Intraspecific interactions, 157-160 Isotopes, stable, 209-218 Israel, Eilat, 31-36, 64-67 J Janes, Stewart W., see Fetz, Trevor W. Juvenile, 340-342 K Kaltenecker, Gregory S., see Yosef, Reuven Keating, Jonathan, see Wayland, Mark Kennedy, Patricia L., see McClaren, Erica L. Kentucky, 344—349 Kershner, Eric L., see Sullivan, Brian L. Kestrel, American, 37-43, 236-243, 286-291, 323-329 Kite, Black, 167-172 Black-winged, 1-7 Red, 167-172 Kleptoparasitism, 91-97, 157-160, 161-163 Kochert, Michael N., see Canavelli, Sonia B. Kornan, Jan, see Kornan, Martin Kornan, Martin, Milos Majda, Metod Macek, and Jan Kor- nan, An unusual case of adoption of a Golden Eagle {Aquila chrysaetos) chick in the Mala Fatra Moun- tains, northwestern Slovakia, 259-260 December 2003 Index to Volume 37 361 Krone, Oliver and Christian Scharnweber, Two White- tailed Sea Eagles {Haliaeetus albicilla) collide with wind generators in northern Germany, 174-176 Kurosawa, Reiko, see Kurosawa, Takashi Kurosawa, Takashi and Reiko Kurosawa, A helper at the nest of Peregrine Falcons in northern Japan, 340- 342 L La Gran Piedra, 330-333 Lacki, Michael J., see Carter, Kristina M. Landscape, 188-197 Langelier, Ken, see Wayland, Mark Lauchstedt, Heidi, see Fetz, Trevor W. Lead, 8-18 poisoning, 8-18 Letters, 84-86, 173-176, 259-261, 350-351 Livezey, Kent B., see Pearson, Robert R. M Maceda, Juan J., see Canavelli, Sonia B. Macek, Metod, see Korhan, Martin Majda, Milos, see Korhan, Martin Male food-delivery call, 198-208 Mallard, 161-163 Mahosa, Santi, Eduardo Mateos, and Vittorio Pedrocchi, Abundance of soaring raptors in the Brazilian Atlan- tic rainforest, 19-30 Margalida, Antoni and Joan Bertran, Interspecific and in- traspecific kleptoparasitic interactions of the Beard- ed Vulture {Gypaetus barbatus) at nesting areas, 157- 160 Margalida, Antoni and Jennifer Boudet, Dynamics and temporal variation in age structure at a communal roost of Egyptian Vultures {Neophron percnopterus) in northeastern Spain, 252-256 Margalida, Antoni, Jordi Canut, and Diego Garcia, Ter- ritory change and nest-site switching in the Bearded Vulture {Gypaetus barbatus), 333-337 Marks, Jeff, a review of Raptors in the New Millennium, Ed. by Reuven Yosef, Michael L. Miller, and David Pepler, 2002, 87 Marks, Jeff, a review of Birds of the Saskatoon Area, Ed. by A.L. Leighton, J. Hay, C.S. Houston, J.F. Roy, and S. Shadick, 2002, 87-88 Marquez-Reyes, Cesar, see Bechard, Marc J. Martell, Mark, see Santana, Freddy Rodriguez Marti, Carl D., a review of North American Owls: Biology and Natural History, Second Edition, by Paul A. Johnsgard, 2002, 352-353 Marti, Carl D., see Oleyar, M. David Martin, Pamela A., see de Sofia, Shane R. Martinez, Jose Antonio, see Zuberogoitia, Inigo Mateos, Eduardo, see Mahosa, Santi McAdie, Malcolm, see Wayland, Mark McClaren, Erica L., Patricia L. Kennedy, and Phillip L. Chapman, Efficacy of male Goshawk food-delivery calls in broadcast surveys on Vancouver Island, 198- 208 Meleagris gallopavo, 337-340 Micohabitat, 188-197 Migration, 63-64, 78-83, 330-333 nocturnal, 64—67 spring, 31-36 Mika, Markus, see Oleyar, M. David Miller, Michael J.R., see Wayland, Mark Milvus migrans, 167-172 milvus, 167-172 Mining, 188-197 Mixed models, 198-208 Monitoring, 344-349 Montevecchi, William A., see Dominguez, Laura Moore, Stan, a review of Cool North Wind: Morley Nel- son’s Life with Birds of Prey, by Stephen Stuebner, 2002, 262-263 Morphology, 44—54 Moser, Brian W. and G. Keith Hilpp, Wintering raptor use of hybrid poplar plantations in northeastern Oregon, 286-291 Mortality, 292-298 Mundy, Peter J., see Hartley, Ronald R. Murphy, Robert K., see Nenneman, Melvin P. N Navarro, Joan, Jose Antonio Sanchez-Zapata, Martina Carrete, Francisco Botella, Andrei Gavrilov, Sergei Sklyarenko, Jose Antonio Donazar, Olga Ceballos, and Fernando Hiraldo, Diet of three sympatric owls in steppe habitats of eastern Kazakhstan, 256-258 Negro, Juan J., see Ferrero, Juan J. Nenneman, Melvin P., Todd A. Grant, Marriah L. Son- dreal, and Robert K. Murphy, Nesting habitat of Cooper’s Hawks in northern Great Plains wood- lands, 246-252 Neophron percnopterus, 252—256 Nest, abundance, 323-329 competition, 333-337 habitat, 135-146, 219-227, 246-252 site, 228-235, 315-322 site characteristics, 323-329 structure, 135-146 tree, 228-235 Nesting areas, 157-160 Newfoundland, 209-218 North Dakota, 246-252 Nyctea scandiaca, 164-166 O Occupancy, 265-276 Oleyar, M. David, Carl D. Marti, and Markus Mika, Ver- tebrate prey in the diet of Flammulated Owls m northern Utah, 244—246 362 Index to Volume 37 VoL. 37, No. 4 Olivo, Cristian, Fall migration of the White-tailed Hawk in central Bolivia, 63-64 Olson, Stores L,, see Suarez, William Ontario, 307-314 Ontiveros, Diego, Nesting distribution, food habits, and conservation of Osprey on Boavista Island (archipel- ago of Cape Verde), 67—70 Oreochromis spp., 292-298 Oregon, 315-322 Osprey, 67-70, 292-298, 307-314, 330-333 Otus Jlammeolus, 244—246 Owl, Barred, northern, 265-276 Boreal, 55-62 Flammulated, 244-246 Great Gray, 315-322 Little, 256-258 Long-eared, 256-258 Pygmy- Ferruginous, 151-157 Northern, 117-124 Snowy, 164-166 Spotted, northern, 265-276 Tawny, 55-62 Ural, 55-62 P Pandion haliaetus, 67-70, 292-298, 307-314, 330-333 Pearson, Robert R. and Kent B. Livezey, Distribution, numbers, and site characteristics of Spotted Owls and Barred Owls in the Cascade Mountains of Wash- ington, 265-276 Pedrocchi, Vittorio, see Mafiosa, Santi Perch site, 151—157 Pesticide poisoning, 8-18 Plumage, 44-54 Point count, 19-30 Poles, 135-146 Polyandry, 342-344 Polygamy, 342-344 Polynyas, 164-166 Poplar hybrid, 286-291 Population, 307-314 Populus spp., 286-291 Predation, 161-163 Predator-prey interaction, 337-340 Prey, 177-187 Productivity, 209-218, 307-314 Puerto Rico, 277-285 Q Quaternary, 71-75 R Radio telemetry, 147-151 Rainforest, Brazilian, 19-30 Raptor, 177-187 Reinhart, Bradley D., see DeVault, Travis L, Rhodes, Olin E., Jr., see DeVault, Travis L. Reproductive output, 98-116 Richness, 19-30 Robertson, Gregory J. and H. Grant Gilchrist, Wintering Snowy Owls feed on sea ducks in the Belcher Islands, Nunavut, Canada, 164—166 Roloff, Gary J., see Hanauska-Brown, Lauri A. Roost, 78—83 communal, 252-256 site, 228-235 Rowell, Petra, Geoffrey L. Holroyd, and Ursula Banasch, The 2000 Canadian Peregrine Falcon Survey, 98-116 S San Clemente Island, 323-329 Sanchez-Zapata, Jose Antonio, see Navarro, Joan Santana, Freddy Rodriguez, Luis Melian Hernandez, Mark Martell, and Keith L. Bildstein, Cuban raptor- migration counts in 2001, 330-333 Santillan, Miguel Angel, see Sarasola, Jose Hernan Sao Paulo State, 19-30 Sara, Maurizio, The colonization of Sicily by the Black Kite {Milvus migrans), 167-172 Sarasola, Jose Hernan, Miguel Angel Santillan, and Max- imiliano Adrian Galmes, Food habits and foraging ecology of American Kestrels in the semiarid forests of central Argentina, 236-243 Scharnweber, Christian, see Krone, Oliver Semiarid forest, 236-243 Sex, 31-36 Sexual size dimorphism, 337-340 Shooting, 292-298 Shrub-steppe, 286-291 Sicily, 167-172 Simon, Melinda M., see Dykstra, Cheryl R. Siskiyou Mountains, 315-.322 Site characteristics, 265-276 Sklyarenko, Sergei, see Navarro, Joan Slovenia, 55-62 Smallwood, John A., Valerie Dudajek, Sivajini Gilchrist, and Mary Anne Smallwood, Vocal development in American Kestrel (Falco sparverius) nestlings, 37-43 Smallwood, Mary Anne, see Smallwood, John A. Smith, Roger N., Stanley H. Anderson, Steven L. Cain, and Jeffrey R. Dunk, Habitat and nest-site use by Red-tailed Hawks in northwestern Wyoming, 219- 227 Snyder, Helen A., see Boal, Clint Sondreal, Marriah L., see Nenneman, Melvin P. Spain 252-256 Sparrowhawk, Levant, 31-36, 64—67 Sperm competition, 1-7 Stability, spatial, 277-285 temporal, 277-285 Status, 44-54, 98-116 December 2003 Index to Volume 37 363 Stephens, Warren L., see DeVault, Travis L. Steppe habitat, 256—258 Strix aluco, 55-62 nebulosa, 315—322 ocddentalis caurina, 265—276 uralensis, 55-62 varia varia, 265-276 Suarez, William, Storrs L. Olson, Red-shouldered Hawk and Aplomado Falcon from quaternary asphalt de- posits in Cuba, 71-75 Sullivan, Brian L., Eric L. Kershner, Sean R Finn, Anne M. Condon, Douglass M. Cooper, and David K. Gar- celon. Nest-site characteristics and linear abundance of cliff-nesting American Kestrels on San Clemente Island, California, 323-329 Survey, 98-116 T Telemetry error, 147-151 Territorial change, 333—337 characteristics, 277-285 signaling, 1-7 Territory, 340-342 Thermals, 63—64 Thompson, Bruce C., see Brubaker, Don L. Tilapia, 292-298 Toxicity, 8—18 Trussell, R. William, see Woodward, Hope D. Tryjanowski, Poitr, see Yosef, Reuven Turkey, Wild, 337-340 U Utah, 78-83, 244-246 V Vertebrate prey, 244-246 Vocalization, 37-43 Vrezec, Al, Breeding density and altitudinal distribution of the Ural, Tawny, and Boreal owls in north Dinaric Alps (central Slovenia), 55—62 Vulture, Bearded, 157—160, 333—337 Egyptian, 252-256 W Washington, 265-276 Wayland, Mark, Laurie K. Wilson, John E. Elliott, Michael J.R. Miller, Trent Bollinger, Malcolm McAdie, Ken Langelier, Jonathan Keating, and Jennifer M.W Froese, Mortality, morbidity, and lead poisoning of eagles in western Canada, 1986—98, 8—18 Wei, Gao, see Deng, Wen-Hong White, Clayton M., see Beingolea, Oscar Wilson, Laurie K., see Wayland, Mark Wilson, Robert and James A. Gessaman, Two large Bald Eagle communal winter roosts in Utah, 78-83 Winter, 78-83 Wintering, 164-166 grounds, 125-134 Wood, Petra Bohall, see Balcerzak, Melissa J. Woodbridge, Brian, see Canavelli, Sonia B. Woodward, Hope D. and R. William Trussell, Lone Har- ris’s Hawk kills Great Blue Heron, 85-86 Wyoming, 219-227 Y Yosef, Reuven, Nocturnal arrival at a roost by migrating Levant Sparrowhawks, 64—67 Yosef, Reuven, Lorenzo Fornasari, Piotr Tryjanowski, Marc J. Bechard, Gregory S. Kaltenecker, and Keith Bildstein, Differential spring migration of adult and juvenile Levant Sparrowhawks {Accipiter brevipes) through Eilat, Israel, 31-36 Z Zaccagnini, Maria E., see Canavelli, Sonia B. Zimbabwe, 44—54 Zuberogoitia, Inigo, Jose Antonio Martinez, Ainara Az- kona, Agurtzane Iraeta, Inaki Castillo, Raul Alonso, and Sonia Hidalgo, Two cases of cooperative breed- ing in Eurasian Hobbies, 342—344 THE JOURNAL OF RAPTOR RESEARCH A QUARTERLY PUBLICATION OF THE RAPTOR RESEARCH FOUNDATION, INC. (Founded 1966) EDITOR IN CHIEF James C. Bednarz James R. Belthoff Clint W. Boat Joan L. Morrison Juan Jose Negro ASSOCIATE EDITORS Marco Restani Ian G. Warrentin Troy I. Wellicome BOOK REVIEW EDITOR Jeffrey S. Marks CONTENTS FOR VOLUME 37, 2003 Number 1 Copulation Behavior of a Potentially Double-brooded Bird of Prey, the Black- winged Kite {ElANUS CAERULEUS). JuanJ. Ferrero,Juan M. Grande, and Juan J. Negro 1 Mortality, Morbidity, and Lead Poisoning of Eagles in Western Canada, 1986-98. Mark Wayland, Laurie K. Wilson, John E. Elliott, Michael J.R. Miller, Trent Bollinger, Malcolm McAdie, Ken Langelier, Jonathan Keating, and Jennifer M.W. Froese 8 Abundance of Soaring Raptors in the Brazilian Atlantic Rainforest. Santi Manosa, Eduardo Mateos, and Vittorio Pedrocchi 19 Differential Spring Migration of Adult and Juvenile Levant Sparrowhawks {AcCIPITER BREWPES) Through Eilat, Israel. Reuven Yosef, Lorenzo Fornasari, Piotr Tryjanowski, Marc J. Bechard, Gregory S. Kaltenecker, and Keith Bildstein 31 Vocal Development in American Kestrel {Falco sparverius) Nestlings. John A. Smallwood, Valerie Dudajek, Sivajini Gilchrist, Mary Anne Smallwood 37 Morphometrics and Status of Ayres’s Hawk-Eagle in Zimbabwe. Ronald r. Hartley and Peter J. Mundy 44 Breeding Density and Altitudinal Distribution of the Ural, Tawny, and Boreal Owls in North Dinaric Alps (Central Slovenia) . ai Vrezec 55 Short Communications Fall Migration of the White-tailed Hawk in Central Bolivia. Cristian oiivo 63 Nocturnal Arrival at a Roost by Migrating Levant Sparrowhawks. Reuven Yosef 64 Nesting Distribution, Food Habits, and Conservation of Osprey on Boavista Island (Archipelago of Cape Verde) . Diego Ontiveros 67 Red-shouldered Hawk and Aplomado Falcon from Quaternary Asphalt Deposits IN Cuba. William Suarez, Storrs L. Olson 71 Subadult and Pale Steppe Eagles Breeding in Mongolia. David h. eihs 75 Two Large Bald Eagle Communal Winter Roosts in Utah. Robert Wilson and James A. Gessaman 78 Letters First Breeding Record for Falco peregrinus in Urban Lima, Peru, with Remarks on the Peruvian Breeding Population. Oscar Beingolea and Clayton M. White 84 Lone Harris’s Hawk Kills Great Blue Heron. Hope D. Woodward and R. William Trussell 85 Book Reviews. Edited by Jeff Marks 87 Manuscript Referees 89 Number 2 Peregrine Falcon Predation on Dunlins and Ducks and Kleptoparasitic Interference from Bald Eagles Wintering at Boundary Bay, British Columbia. Dick Dekker 91 The 2000 Canadian Peregrine Falcon Survey. Petra Rowell, Geoffrey l. Hoiroyd, and Ursula Banasch 98 Breeding Season Habitat Use and Ecology of Male Northern Pygmy-Owls. Alan R. Giese and Eric D. Forsman 117 Habitat Use by Swainson’s Hawks on Their Austral Wintering Grounds in Argentina. Sonia B. Canavelli, Marc J. Bechard, Brian Woodbridge, Michael N. Kochert, Juan J. Maceda, and Maria E. Zaccagnini 125 Raptor and Chihuahuan Raven Nesting on Decommissioned Telephone-line Poles in the Northern Chihuahuan Desert. Don L. Brubaker, Kathleen L. Bmbaker, and Bruce C. Thompson 135 Short Communications Aerial Telemetry Accuracy in a Eorested Landscape. Travis l. DeVauit, Warren l, Stephens, Bradley D. Reinhart, Olin E. Rhodes, Jr., and I. Lehr Brisbin, Jr. 147 Perch-site Selection and Spatial Use by Cactus Ferruginous Pygmy-Owls in South- central Arizona. Aaron D. Flesch 151 Interspecific and Intraspecific Kleptoparasitic Interactions of the Bearded Vulture ( GyPAETUS BARBATUS) at Nesting Areas. Antoni Margalida and Joan Bertran 157 Gyrfalcon Predation on Mallards and the Interaction of Bald Eagles Wintering IN Central Alberta. Dick Dekker and Gordon Court 161 Wintering Snowy Owls Feed on Sea Ducks in the Belcher Islands, Nunavut, Canada. Gregory J. Robertson and H. Grant Gilchrist 164 The Colonization of Sicily by the Black Kite {Milvus migrans). Maurizio Sara 167 Letters Gyrfalcon Color Variation. Ian Flann 173 Two White-tailed Sea Eagles {Haliaeetvs ALBiaiLA) Collide with Wind Generators in Northern Germany. Oliver Krone and Christian Scharnweber 174 Talon-locking in the Red-tailed Hawk. Robert W. Dickerman 176 Number 3 Behavior and Prey of Nesting Red-shouldered Hawks in Southwestern Ohio. Cheryl R. Dykstra, Jeffrey L. Hays, Melinda M. Simon, and E. Bernard Daniel 177 Red-shouldered Hawk {Buteo lineatus) Abundance and Habitat in a Reclaimed Mine Landscape. Melissa J. Balcerzak and Petra Bohall Wood 188 Efficacy of Male Goshawk Food-delivery Calls in Broadcast Surveys on Vancouver Island. Erica L. McClaren, Patricia L. Kennedy, and Phillip L. Chapman 198 Reproductive Success, Environmental Contaminants, and Trophic Status of Nesting Bald Eagles in Eastern Newfoundland, Canada. Laura Dominguez, William Montevecchi, Neil M. Burgess, Joe Brazil, and Keith A. Hobson 209 Habitat and Nest-site Use by Red-tailed Hawks in Northwestern Wyoming. Roger N. Smith, Stanley H. Anderson, Steven L. Cain, and Jeffrey R. Dunk 219 Nest and Roost Habitat Characteristics of the Grey-faced Buzzard in Northeastern China. Wen-Hong Deng, Gao Wei, and Zheng Guang-Mei 228 Food Habits and Foraging Ecology of American Kestrels in the Semiarid Forests of Central Argentina. Jose Heman Sarasola, Miguel Angel Santillan, and Maximiliano Adrian Galmes 236 Short Communications Vertebrate Prey in the Diet of Flammulated Owls in Northern Utah. M. David Oleyar, Carl D. Marti, and Markus Mika 244 Nesting Habitat of Cooper’s Hawks in Northern Great Plains Woodlands. Melvin P. Nenneman, Todd A. Grant, Marriah L. Sondreal, and Robert K. Murphy 246 Dynamics and Temporal Variation in Age Structure at a Communal Roost of Egyptian Vultures {Neophron percnopterus) in Northeastern Spain. Antoni Margalida and Jennifer Boudet 252 Diet of Three Sympatric Owls in Steppe Habitats of Eastern Kazakhstan. Joan Navarro, Jose Antonio Sanchez-Zapata, Martina Carrete, Francisco Botella, Andrei Gavrilov, Sergei Sklyarenko, Jose Antonio Donazar, Olga Ceballos, and Fernando Hiraldo 256 Letters An Unusual Case of Adoption of a Golden Eagle {Aquila chrysaetos) Chick in the MaiA Fatra Mountains, Northwestern Slovakia. Martin Kornan, Milos Majda, Metod Macek, and Jan Kornan 259 SiBLICIDE AND CANNIBALISM IN THE BOOTED EaGLE {HiERAAETUS PENNATUS) IN THE TIETAR VaLLEY, Central Spain. Ignacio S. Garda Dios 261 Book Review. Edited by Jeffrey S. Marks 262 Number 4 Distribution, Numbers, and Site Characteristics of Spotted Owls and Barred Owls in the Cascade Mountains of Washington. Robert r. Pearson and Kent B. Livezey 265 Temporal and Spatial Stability of Red-tailed Hawk Territories in the Luquillo Experimental Forest, Puerto Rico, cunt w. Boai, Helen a. Snyder, Brent d. Bibles, and Tracy S. Estabrook 277 Wintering Raptor Use of Hybrid Poplar Plantations in Northeastern Oregon. Brian W. Moser and G. Keith Hilpp 286 Mortality of Wintering Ospre\s and Other Birds at Aquaculture Facilities in Colombia. Marc J. Bechard and Cesar Marquez-Reyes 292 Blood Chemistry, Cytology, and Body Condition in Adult Northern Goshawks {ACCIPITER GENTILIS) . Lauri A. Hanauska-Brown, Alfred M. Dufty, Jr., and GaryJ. Roloff. 299 Productivity AND Population Trends of Osprey in the Kawartha Lakes Region, Ontario, 1978—2001. Shane R. de Solla, Pamela a. Martin, Peter J. Ewins, and Michael E. Barker 307 Habitat Characteristics of Great Gray Owl Sites in the Siskiyou Mountains of Southwestern Oregon. Trevor W. Fetz, Stewart W. janes, and Heidi Lauchstedt 315 Nest-site Characteristics and Linear Abundance of Cliff-nesting American Kestrels on San Clemente Island, California. Brian l. Suiiivan, Eric l. Kershner, Sean P. Finn, Anne M. Condon, Douglass M. Cooper, and David K. Garcelon 323 Short Communications Cuban Raptor-migration Counts in 2001. Freddy Rodriguez Santana, Luis Melian Hernandez, Mark Martell, and Keith L. Bildstein 330 Territory Change and Nest-site Switching in the Bearded Vulture {Gypaetus BARBATUS) . Antoni Margalida, Jordi Canut, and Diego Garcia 333 Immature Northern Goshawk Captures, Kills, and Feeds on Adult-sized Wild Turkey. Gregory H. Golet, Henry T. Golet, and Albie M. Colton 337 A Helper at the Nest of Peregrine Falcons in Northern Japan. Takashi Kurosawa and Reiko Kurosawa 340 Two Cases of Cooperative Breeding in Eurasian Hobbies, inigo Zuberogoitia, Jose Antonio Martinez, Ainara Azkona, Agurtzane Iraeta, Ihaki Castillo, Raul Alonso, and Sonia Hidalgo 342 Food Habits of Peregrine Falcons in Kentucky. Kristina m. Carter, Michael j. Lacki, Matthew R. Dzialak, Laura S. Burford, and Ronald O. Bethany 344 Letter Harris’s Hawk Dies After Colliding with Motor Vehicle. Jennifer O. Coulson and Thomas D. Coulson 350 Book Review. Edited by Jeffrey S. Marks 352 Information for Contributors 354 Index to Volume 37 358 A Telemetry Receiver Designed with The Researcher in Mind What you've been waiting for! Fin.iUy. A highly Mrnsitive 999 channel synthesized telemetry receiver that weighs less than 13 ounces. Is completely user programmable and offers variable scan rates over all frequencies. For each animal being tracked, the large ICD display provides not only the frequency (to lOOMz) and channel number, but also a 7 character alphanumeric comment field and a digital signal strength meter. Stop carrying receivers that are the size of a lunch bo* or cost over SlSOO. The features and performance of the new R-lOOO pocket sized telemetry receiver wilt impress you. and the price will convince you. Othtr Features Inctude: • factory tuned to any AMUz wide segment in the 148 174MHz Band • Very high sensitivity of -148dBm to -150dBm • Illuminated display and keypad for use m low light or darkness • User selectable scan rates from 1-30 seconds in 1 second steps • Rechargeable batteries operate the receiver for 12 hours and JLX can be replaced with standard AA Alkaline batteries in the field. Both 12vdc and llOvac chargers are included. 6.r 2 . 6 ' 1.5' (15.Scm) high, (6.6cm) wide, (3.8cm) deep. • 3 year warranty • 1 day delivery $ 695.00 Rlease specify desired 4MHz wide segment in the 148-174HMZ band Visit our website for complete specifications, operating manual and information on tin* R 1000 Of call our loll ftff number to order yuur receiver ruiw. Try the New R‘1000 and You*U Be Impressed! COMMUNICATIONS SPECIALISTS, INC. ( A S A .*] ! « orders:8oo-722-246o Phone:434-263-4842 fax:434-263-4842 Buteo Books is the largest retailer of Ornithology books in North America^ with over 2^000 in-print titles in stock. A FEW BOOKS ON BIRDS OF PREY AND FALCONRY Handbook of the Birds of the World. The first seven volumes of this projected sixteen-volume work are available for $185 each. Volume 2 includes diurnal raptors, Volume 3 includes owls. Volume 8, the first to cover Passerines, due Summer 2003. A Photographic Guide to North American Raptors. Wheeler & Clark. $24.95 (New Eastern and Western editions by Wheeler are due September 2003.) Ecology and Conservation of Owls. Ian Newton. $95 Birds of Prey: Health and Disease. Third Edition. Cooper. $96.95 North American Falconry and Hunting Hawks. Beebe & Webster. $55 RARE AND OUT-OF-PRINT ORNITHOLOGY Buteo Books has a wide selection of used and out-of-print titles available. Our stock changes daily, so call to check availability of that scarce title you're seeking. BIRDS OF NORTH AMERICA SERIES In 1992 the American Ornithologists' Union, in partnership with the Academy of Natural Sciences of Philadelphia, undertook the publication of species accounts for each of the more than 700 species which breed in the United States and Canada. These illustrated reviews provide comprehensive summaries of the current knowledge of each species, with range maps and an extensive list of references. Buteo Books is pleased to offer these individual species accounts for $7.50 each. All 716 profiles are listed in taxonomic order on our website (conforming with the Seventh Edition of the A.O.U. Check-list of North American Birds). Singles may be ordered by contacting Buteo Books. Shipping and handling is $4 for the first profile and $1 for each additional profile to a maximum of $10 per order. Visit our website for more information: www.buteobooks.com allen@buteobooks . com 3130 Laurel Road; Shipman, VA 22971; USA 2004 ANNUAL MEETING The Raptor Research Eoundation, Inc. 2004 annual meeting will be held on 9-13 November 2004 in Bakersfield, California. For information about the meeting see the following website: bttp:/ /www. calhawkingdub.org/field_meet/34th_annual/ or contact Rick Holderman (parabuteol@cox.net). Persons interested in predatory birds are invited to join The Raptor Research Foundation, Inc. Send requests for information concerning membership, subscriptions, special publications, or change of address to OSNA, P.O. Box 1897, Lawrence, KS 66044-8897, U.S.A. The Jou7'nal of Raptor Research (ISSN 0892-1016) is published quarterly and available to individuals for $33.00 per year and to libraries and institutions for $50.00 per year from The Raptor Research Foundation, Inc., 14377 1 17th Street South, Hastings, Minnesota 55033, U.S.A. (Add $3 for destinations outside of the continental United States.) Periodicals postage paid at Hastings, Minnesota, and additional mailing offices. POSTMASTER; Send address changes to The Journal of Raptor Research, OSNA, P.O. Box 1897, Fawrence, KS 66044-8897, U.S.A. Printed by Allen Press, Inc,, Lawrence, Kansas, U.S.A. Copyright 2003 by The Raptor Research Foundation, Inc. Printed in U.S.A. @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Raptor Research Foundation, Inc., Awards Lifetime Achievement Awards The Tom Cade Award recognizes an individual who has made significant advances in the area of captive prop- agation and reintroduction of raptors. Nomination packets can be submitted at any time. Contact: Brian Walton, Predatory Bird Research Group, Long Marine Laboratory, University of California, Santa Cruz, CA 95064 U.S.A.; tel. 408-459-2466; e-mail: walton@cats.ucsc.edu. The Fran and Frederick Hamerstrom Award recognizes an individual who has contributed significantly to the understanding of raptor ecology and natural history. Nomination packets can he submitted at any time. Con- tact; Dr. Clint Boal, Texas Cooperative Fish and Wildlife Research Unit, BRD/USGS, Texas Tech University, 15th Street & Boston, Ag Science Bldg., Room 218, Lubbock TX 79409-2120 U.S.A.; tel. (806) 742-2851; e-mail; cboal@ttacs.ttu.edu. Student Recognition and Travel Assistance Awards The James R. Koplin Travel Award is given to a student who is the senior author and presenter of a paper or poster to be presented at the RRF annual meeting for which travel funds are requested. Contact: Dr. Patricia A. Hall, 5937 E. Abbey Rd. Flagstaff, AZ 86004 U.S.A.; tel. 520-526-6222; e-mail: pah@spmce.for.nau.edu. Application Deadline: due date for meeting abstract. The W ill iam C. Andersen Memorial Award is given to the students who are senior authors and presenters of the best student oral and poster presentation at the annual RRF meeting. Contact: Laurie Goodrich, Hawk Mountain Sanctuary, 1700 Hawk Mountain Road, Kempton, PA 19529 U.S.A.; tel. 610-756-6961; email: goodrich@hawkmountain.org. Application Deadline; due date for meeting abstract; no special application is needed. Grants For each of the following grants, complete applications must be submitted to the contact person indicated by 15 February. Recipients will be notified by 15 April. The Dean Amadon Grant for $200-400 is designed to assist persons working in the area of distribution and sys- tematics (taxonomy) of raptors. Contact: Dr. Carole Griffiths, 251 Martling Ave., Tarrytown, NY 10591 U.S.A.; tel. 914-631-2911; e-mail: cgriff@liu.edu. The Stephen R. TuUy Memorial Grant for $500 is given to support research, management, and conservation of raptors, especially to students and amateurs with limited access to alternative funding. Contact; Dr. Kim Titus, Alaska Department of Fish and Game, Division of Wildlife Conservation, P.O. Box 240020, Douglas, AK 99824 U.S.A.; e-mail: kimt@fishgame. state. ak.us. The Leslie Brown Memorial Grant for up to $1,000 to support research and/or dissemination of information on birds of prey, especially to proposals concerning African raptors. Contact: Dr. Jeffrey L. Lincer, 9251 Golondrina Dr., La Mesa, CA 91941 U.S.A.; e-mail: jefflincer@tns.net.