j. RaptorRes. 30(2):93-98 ¸ 1996 The Raptor ResearchFoundation, Inc.

A COMPARISON OF BEHAVIOR AND SUCCESS RATES OF MERLINS AND PEREGRINE FALCONS WHEN HUNTING DUNLINS IN TWO COASTAL HABITATS

JOSEPHB. BUCHANAN CascadiaResearch Collective, 218• WestFourth Avenue, Waterstreet Building, Olympia,WA 98501 U.S.A. and WildlifeManagement Program, Washington Department of Fish and Wildlife,600 CapitolWay North, Olympia,WA 98501 U.S.A

ABSTRACT.--Thehunting behavior and successof raptors have been linked to prey availability,habitat conditions,and competition.In this study,I comparedthe behaviorand hunting efficiencyof wintering merlins (Palcocolumbarius) and peregrine falcons(E peregrinus) that hunted dunlins (Calidrisalpina) at coastalestuarine and beachhabitats in Washingtonto determine whether hunting efficiencywas related to differencesin habitat. Flocksof up to 15,000 dunlins moved 2-15 km from estuarinesites to roost and forageat beachesduring diurnal high tides.Both falconspecies regularly attacked flocks of dunlins in each habitat.The successrates of huntingflights for both species(merlin: 7.8%, peregrinefalcon: 12.5%) were significantlylower at beachesthan previouslyreported for estuaries.Dunlins at the beach habitatroosted in flocksnear water'sedge and avoidedfalcons by flying out over the oceanwhere flocks engagedin synchronizedflight in the troughsbetween waves. Most capture attempts by falcons occurred over water even though the likelihood of successfulcapture was lower there. Neither age of the hunting raptor (peregrinefalcon only) nor relativedensity of potential kleptoparasitesinfluenced hunting suc- cess.Several hypotheses are presented to explain differencesin hunting efficiency between the two habitats.

KEYWORDS: dunlin; Falco columbarius;Falco peregrinus;merlin; peregrine falcon; predation; Washington; winter:

Comparacitn de conductasy tasasde &xito de Palcocolumbarius y Falco pereg•nus en la caza de Calidris alpinaen dos hfibitatscosteros RESUMEN.--Tantola conductade caza coxnoel &xito de rapacesha sido ligada a la disponibilidadde presas,condiciones del h•tbitaty competencia.En esteestudio, cornpar& la conductay la eficienciade caza de Palcocolumbarius y Falco peregrinus que capturaron Calidrisalpina en estuariosy playas,hfibitat costerosde Washington,con el fin de determinarsi la eficienciade cazaestaba relacionada a diferencias en hfibitat. Durante la marea alta diurna, bandadasde C. alpina, sobre los 15.000 individuos,se movian de dos a 15 km desdeel estuarioa sitiosde descansoy forrajeo en playas.Las dos especiesde halcones atacaronregularmente a estasbandadas (C. alpina) en ambos tipos de hfibitats.La tasa de &xito en capturaal vuelopara ambas especies (E columbarius= 7.8%, E pereg•nus= 12.5%)fue significativamente milsbaja en playasqueen estuariospreviamente reportadas. Calidris alpina descans6 en hfibitat de playas cercanosal agua y evitaron el ataque de los halconesvolando hacia el mar en sincronizadasbandadas. I,a mayoria de los intentos de captura por halconesocurri6 sobre el agua, donde la probabilidad de •xim era la milshaja. Tanto la edad del rapaz cazador(s61o E pereKrinus) como la densidadde potenciales kleptoparfisitosno influenciaron el •xito de captura. Presentovarias hipttesis para explicar las difer- encias en eficiencia de caza entre los dos hfibitats. [Traducci6n de Ivan Lazo]

Studies of raptor fbraging ecology have identi- habitatswhere caloric intake is higher. These find- fied relationshipsbetween habitat use and factors ings are consistent with the ideal free model of such as vegetation structure and prey abundance habitat selection (Fretwell and Lucas 1970) which (Wakeley 1978a, 1978b, 1979, Baker and Brooks holds that marginal habitatswill be used by certmn 1981, Bechard 1982, Toland 1987, Preston 1990). individualsin a population to meet some or all life They show that raptors preferentially forage in reqnisites.Few studies have attempted to describe

93 94 BUCHANAN VOL. 30, No. 2 foraging efficiencyand examine its relationshipto as high as 40,000-70,000 birds, respectively(Paulson differing habitat conditions, prey populations,or 1993, Evensonand Buchanan,unpubl. data). Although some dunlins remained within these estuariesat high prey behavior (Swenson1979, Bildstein1987, To- tide, quite large numbers moved to the outer beaches, land 1987). densitiesof roostingdunlins averaged >400 birds/km on In western Washington, the merlin (Falcocolum- beaches(Buchanan 1992). Roostingand in-flight flocks barius)and peregrine falcon (E peregrinus)are ma- >2000 dunlins were observedin both habitats (Brennan et al. 1985, Buchanan 1992). jor predators of the dunlin (Calidris alpina; Bu- chanan et al. 1986, 1988, Dobler and Spencer METHODS 1989), the most abundant winter shorebird in the Data Collection. Field work wasconducted on 38 days region (Brennan et al. 1985, Evensonand Buchan- between Nov-Feb 1983-90 and Nov-Mar 1993-94. Ob- an 1995). The behavior of hunting falcons has servationsat estuarieswere made in 1979-90 (Buchanan been describedfor flights directed at roosting or et al. 1986, 1988, unpubl. data). I traveled a cumulative foraging flocks of dunlins in estuaries (Buchanan total of 894 km of shorelineconducting shorebird counts and observingfalcons. Each beach was visited at least et al. 1986, 1988). Large populations of wintering once each winter with the exception of 1993-94, when dunlins occur at two coastal estuaries (Buchanan South Beach was not visited. Field work was restricted to and Evenson, unpubl. data); many of these birds a period 3 hr before and after high tide; high tidesdur- make short flights each day during high tide to ing field visitsranged between2.3 and 3.2 m. High tides outer beach roost sites (Buchanan 1992). Individ- >2.3 m inundated all tidal flats in the adjacentestuaries under most conditions. The majority of diurnal high ual falcons track these movements and hunt shore- tides during Nov-Feb were >2.3 m and ranged as high birds in both habitats. To determine whether hunt- as 3.2 m. ing efficiency was related to habitat conditions, I I made most observationsfrom a vehicle, using 7x 35 observed the behavior and efficiency of merlins binocularsand a 25x spottingscope. Certain sectionsof beach were coveredby foot. When falcons hunted shore- and peregrine falcons hunting dunlins at outer birds I recorded movementand positionof the flock dur- beachesto compare with earlier data collected at ing attack, type of hunting flight by the falcon, and the estuariesin western Washington (Buchanan et al. number and type of capture attemptsmade. 1986, 1988). Definitions and descriptionsof falcon hunting behav- ior are provided by Buchananet al. (1988), and are de- SrUDY AP,EA scribedhere briefly.A huntingflight is a perch-to-perch flight that includesone or more captureattempts. A cap- Observations were made at three beaches on the outer ture attempt is defined as an attempt to seize or knock coastof Washingtonstate: Copalis-North Beach (25 km down prey. Exploratory"feints" are not consideredcap- •n length), South Beach (23 kin), and Long Beach (37 ture attempts.Hunting methodsused by merlinsand per- kin). The beachesare contiguous,relatively fiat expanses egrine falconsincluded stoops,nearly vertical,rapid de- of sand backed by low dunes. Beach width during high scentstoward flocks or singlebirds; stealth approach, rap- tide rangesfrom 0-30 m, dependingon locationand tide id, low (usually<2 m) flights towardroosting flocks; and height.The high-tidezone is characterizedby opensand, horizontal chases,the pursuit of either flocks or single varying amounts of logs, and tidal debris. Most of the birds, often after another hunting method failed. length of thesebeaches occurs along two peninsulassit- Data Analysis.To allowcomparison with other studies, uated perpendicularto the openingsof GraysHarbor I calculatedthe proportion of hunting flights and cap- and Willapa Bay;only a 13 km length of South Beachis ture attemptsthat were successful.In certain cases,I ob- not directlyacross the peninsulafrom an intertidal area. served a sequenceof hunting flights involving a single The peninsulasare 1.5-3 km wide and are dominatedon falcon. Although sequentialobservations are not statisti- the windwardside by low dunes.Dune vegetationis pre- cally independent events (Beal and Khamis 1990), they dominantlyEuropean beach grass(Ammophila arenaria) can be used to increasesample size (Hejl et al. 1990), and wax myrtle (Myrica californica).In some areasforests particularlywhen samplesare difficult to obtain, or to of Sitka spruce (Piceasitchensis) and lodgepolepine (Pi- minimize samplingerror associatedwith singlesampling nus contorta)occur within 150 m of the dunes. (Morrison 1984). Recently, Leger and Didrichsons Observationsof falcon hunting behavior at estuaries (1994) found that use of pooled observationsdid not (Buchananet al. 1986, 1988) were made primarily at four causebias if intra-subjectvariance exceeded between-sub- sites:Samish Bay in north Puget Sound, NisquallyRiver ject varianceor subjectsample sizes were the same. delta and Totten Inlet in south Puget Sound, and Grays I evaluated the dataset to determine whether inclusion Harbor. These sites are described in Brennan et al. of all sequentialobservations in the analysiswas appro- (1984, 1985). Observations at beach and estuarine sites priate. First, I randomly selectedthree hunting flights were made under a varietyof weather conditionstypical per falcon and recordedthe number of captureattempts of the mild, wet winters in the region. per flight and calculatedboth intra-subjectvariance and Winter populationsof dunlinsat three of the estuaries between-subjectvariance (Leger and Didrichsons1994) ranged in size from 2000-13,000 birds (Brennan et al. The calculatedratio of intra-subjectvariance to between- 1985). Counts at Grays Harbor and Willapa Bay ranged subjectvariance was 1:0.95 (i.e., intra-subjectvariance ex- 1996 F^LCONPREDATION ON DUNLINS 95 ceeded between-subjectvariance). I next calculatedthe cended and descended in an undulating motion number of different merlins and peregrine falconsseen with the passingof each cresting wave. All evasive hunting in each habitat. For this calculationI assumed that birds tallied between yearswere different birds. The flights at beachesoccurred within 20 m of the wa- numbers of fitIcons observed hunting in each habitat ter, and in most cases the bottom of the flock was were nearlyidentical. At least 17 and 23 merlinshunted <1 m above water. shorebirdsat beachesand estuaries,respectively, and 11 Behavior of Falcons.All 75 hunting flights by different peregrine falcons were observed hunting in falcons at beaches were directed at dunlins. Hunt- each habitat. Finally, observationsat beach siteswere spa- tially and temporally independent from those made at ing flights were directed at single birds and flocks estuaries.For these reasons,I felt justified to use all se- ranging in sizeup to 12,000 dunlins. All three cap- quential observations. tures made by peregrine falcons were brief (1-2 To determine whether falcons hunted with similar rel- capture attempts/flight) pursuits of solitary dun- ative frequenciesin the two habitats,I compared mean •ndicesof hunting activity (number of flights/hr/yr) at lins that had split (or were forced) away from each habitatusing a Mann-Whitneytest. I combinedboth flocks; captures made by merlins were of birds speciesfor this analysisbecause the only two estuarine taken directly from in-flight flocks of about 4000 sitesthat supported peregrine falconswere not visited all and 60 dunlins. Both speciesoccasionally contin- years (Buchanan et al. 1986). To test the null hypothesis that there was no difference in the efficiency of falcons ued to chase birds they isolated from a flock, but hunting dunlins in different habitats, I assessedhunting no such extended pursuitswere successful.Two of efficiency in three ways. First, I compared successrates sevenprey capturesoccurred over water. by hunting flight and by individual capture attempt for Falconswere equally likely to be observed hunt- merlins and peregrine falcons at beachesand estuaries ing in either habitat. The index of hunting inci- with a 2x2 chi-square contingency analysis(Zar 1984), with a correction for continuity (df = 1). Second, I com- dence (hunts/hr/yr) was the same at beach (2 = pared the number of capture attemptsper hunting flight 0.64, SD = 0.89) and estuarine sites (• = 0.55, SD in each habitat using the Mann-Whitney test. Finally, I = 0.44; U = 46, P > 0.20). compared the number of capture attempts made per I observed hunting flights by merlins (51) and flight over land and over water for both speciesusing the Wilcoxon test. peregrine falcons (24) directed at flocks of dunlins I assessedthe effectsof two factors, age of falconsand at beach habitat and found the success rates for presenceof kleptoparasites,on hunting efficiency.Age is hunting flights at beach siteswere significantlylow- known to influence hunting efficiencyin birds and youn- er than at estuariesfor both species(merlin: 8 vs. ger, inexperienced birds generally are lesssuccessful at 23%, X9• = 4.17, P = 0.043; peregrine falcon: 13 securingprey than adults (Burger 1988). My samplesize was too small to fully evaluate the relationship between vs. 47%, X=c= 4.00, P = 0.046). The successrates age and hunting success.However, for the peregrine fal- for capture attemptswere alsolower at beach sites, con I used chi-square analysisto compare the propor- but the differences were not significantlydifferent tions of hunting flights involvingsubadults in each hab- itat to determine whether differences in success rate (merlin: N = 176 captureattempts, 2 vs.6%, Xe•= might be explained simplyby the age ratio of the birds 2.42, P = 0.13; peregrine falcon: N = 52 capture I observedhunting in the two habitats.This analysiswas attempts,6 vs. 15%, X=•= 2.26, P = 0.15). not conducted for merlins becauseit was not alwayspos- The number of capture attempts per flight at sible to determine age of hunting merlins. Becausepre- beach sites (merlin: • = 3.45, SD = 3.52; peregrine vious studiesfound that merlin hunting behavior varied falcon: • = 3.47, SD = 2.92) were the same as at significantlyin the presenceof kleptoparasites(Buchan- an 1988), I attempted to determine whether the pres- estuarine sites (merlin: 2 = 3.81, SD = 4.77, Z = ence of known potential kleptoparasites (Brockmann 0.04, P > 0.50; peregrine falcon: x = 2.75, SD = and Barnard 1979) influenced the successrates of hunt- 2.4; U = 152.5, P > 0.20). Both falcons made fkwer mg flights in the two habitats. For this analysis I com- capture attempts per flight over land (merlin: x = pared mean density indices (birds/hr/yr) of potential kleptoparasitesat beach and estuarine sitesusing a one- 0.69, SD = 0.68; peregrine falcon: 2 = 0.5, SD = tailed t-test. 1.06) than over water (merlin: 2 = 2.78, SD = 3.6, 72 = 248, P < 0.001' peregrine fhlcon: • = 2.5, SD RESULTS = 2.25, T_ = 3, P < 0.001). All fbur prey captures Behavior of Dunlins at Beaches. Dunlins re- by merlins occurred over the beach. Two of three sponded to attacks from falcons by flying directly captures of prey by peregrine fialconsoccurred over the ocean where flocks engaged in rapid, over the water when single birds were captured highly synchronizedevasive movements. All pred- awayfrom flocks;the third capture occurred when ator evasionflights occurred over the water. Flocks a single bird was taken above the beach afier leav- nearly alwaysflew very low over the water, and as- ing a flock out over the water. The successrates of 96 BUCHANAN VOL. 30, No. 2 capture attempts made over beaches by merlins (a gyrfalconattempted to take prey from a pere- and peregrinefalcons were 11.4 and 8.3%, respec- grine falcon immediatelyafter capture). tively.Only 5% of capture attemptsby peregrine DISCUSSION falcons over water were successful. When attackinga flock over the beach,both spe- Merlins and peregrine falcons hunted less effi- ciesalways used low stealthapproaches above land; ciently, in terms of the successrate of hunting there were no initial approaches over water. In flights, at beaches(8 and 13%, respectively)than some cases,the flock saw an approaching falcon at estuarinesites (23 and 47%, respectively).Both •250 m away and moved to a position over the specieshad similar successrates for capture at- breakers where much of the hunting activity oc- temptsat beach sites(2 and 6%, respectively)and curred. In such cases, the falcon either continued estuaries(6 and 15%, respectively)but peregrine the attack directly or circled to a position above falconswere more successfulthan merlinsat cap- the breakers(merlins: 20-50 m, peregrine falcons: turing dunlins. Although peregrine falconsare of- 40-80 m), from where they initiated stoopsor dis- ten far more successfulat securingprey (Ratcliffe continued the attack. For merlins, 17 flights oc- 1980), relativelylow ratesof successhave been doc- curred exclusivelyover the beach, 12 were initiated umented (Bertochi et al. 1984, seereview by Roalk- over the beach and moved to water (e.g., capture vam 1985). The variation in rates of successful attemptswere made abovebeach and water during hunting flights by peregrine falcons and merlins a hunting flight), and 22 occurredexclusively over have been attributed to factorssuch as age, expe- water (e.g., the capture attempt occurred over wa- rience, degree of intent, and energy requirements ter). This pattern of hunting location differed (Ratcliffe 1980, Treleaven 1980, Sodhi et al. 1993); slightly for peregrine falcons, where three flights factorsassociated with the hunting successrates of occurred exclusivelyover beach, four were initiat- other raptor speciesinclude the behaviorand vul- ed over beach and moved to water, and 17 oc- nerability of prey and vegetationstructure (Swen- son 1979, Bildstein 1987, Toland 1987). curred exclusivelyover water (2X3 contingency test; X2 = 5.42, P = 0.07). Low attacksoccurred in Solitary shorebirds can be more vulnerable to 38 (75%) and 10 (42%) of the hunting flightsby predation than birds in flocks(Kus 1985). Species reliant upon synchronizedflocking when threat- merlins and peregrine falcons,respectively; most ened by predatorsperhaps benefit from this be- low approaches were stealth flights, but some havior by reducing the probability of predation flights over water (merlin = 10, peregrine falcon (Kus 1985), by confusing the predator (Davis = 5) also included low pursuit of singlebirds. All 1980). In this studyI found that five of sevenprey stoopsby both specieswere made over water and captures were of single birds isolated from flocks. occurred in most flights that were exclusively Most captures occurred above beacheswith only abovewater (merlin: 16 of 22, 73%, peregrinefal- two captures occurring over water. There were no con: 15 of 17, 88%). successfulcaptures made through attackson large Influences on Hunting Efficiency. Nine of 24 flocks over water. This was in sufikingcontrast to hunting flights by peregrine falcons at beaches hunting behavior at estuarieswhere 43 and 32% (38%) were made by subadultbirds. This is similar of all prey capturesby peregrine falconsand mer- to the proportion of flights by subadultsobserved lins, respectively,occurred during stoopsat flocks at estuaries (33%; X2• = 0.08, P = 0.78). of dunlins (Buchanan et al. 1986, 1988). In both I observedsix raptor speciesin the two habitats habitats, attacks at flocks resulted in single birds that are known to steal prey from other raptors: becoming isolated from the flock and these were bald eagle (Haliaeetusleucocephalus), northern har- most successfulwhen dunlins were not flying over rier (Circuscyaneus), red-tailed hawk (Buteojamai- wave s. censis),rough-legged hawk (B. lagopus),gyrfalcon Shorebirdsare known to evadepredators by fly- (F rusticolus),and peregrine falcon. As a group, ing over water (Hunt et al. 1975, Bertochi et al. these specieswere observed at a much lower fre- 1984, Boyce 1985, Buchanan et al. 1988). The quency (birds/hr/yr) at beaches (/= 0.34, SD = height of dunlin evasiveflights I observedat beach- 0.22) than at estuaries (i = 1.42, SD = 0.4; t = 6.6, es waslower than at estuaries(e.g., 50 m; Buchan- d•: = 1, 14, P < 0.0005). I observedonly one in- an et al. 1988), suggestingan adaptiveadvantage •,nce of attempted kleptoparasitismat beach sites of dunlins flying low over the water to evade fal- JUNE1996 FALCONPREDATION ON DUNLINS 97 cons.There are severalpossible reasons why shore- should hunt in areaswhere prey is abundant and birds evade predators by flying over water and why available enough to offset costsassociated with for- falcons are less effective when hunting shorebirds aging. Falcons may hunt in beach habitat, where in beach habitat. First, shorebirds are capable of they are less efficient predators, in order to meet safelylanding in water to escapecapture (Hunt et their high energetic costs.A merlin, for example, al. 1975, Buchanan et al. 1991). Although both requiresabout 70-75 g food each day (Page and peregrine falcons and merlins are known to re- Whitacre 1975), and in westernWashington would trieve floating or swimmingbirds from water (Dek- need to consume about 3 dunlins (Brennan et al. ker 1980, Boyce 1985, Buchanan et al. 1991), they 1984) or similar:sizedprey every 9 daysto meet th•s may not be able to do so in turbulent water (e.g., intake requirement. Owing to the short daylength the breaker zone). If falcons are reluctant to re- and very high diurnal tides during winter, shore- trieve such birds, perhapsthey must attempt more birds may spend as much or more time at beach difficult captures of prey in mid-air. This is an un- sites than at estuarieson certain days. Therefore, likely explanation becausefalcons typicallysecure a merlin may need to hunt shorebirdsin lessop- shorebirdsin flight. Second, despite the fact that timal conditions to meet its average daily energy merlins and peregrine falconsoccasionally hit the requirement, unless other prey are available. Th•s water at the terminus of stoops during hunting expended effort may be offset by the greater suc- flights (three observationsfor each speciesat es- cessrate for hunting flights at estuaries, and the tuaries;J.B. Buchanan, unpubl. data), the risk of fact that falcons can more easilytrack flock move- hitting a wave during a stoop, or of being inun- ment between habitats (e.g., search time is re- dated by a crestingwave may influence a falcon's duced). However, it is unknown whether the lower ability to capture prey so close to the water. It is successrate of hunting flights at beaches repre- difficult to assessthe potential importance of this sents a significant physiological stress to falcons factor. Third, falcons may simply be confusedby and whether other potentially suitable prey are the contrasting movements of individuals within available during the periods when dunlins are the flock relative to the movement of the waves roosting at outer beaches.These issuesmust be ad- and may have difficulty trackingtarget birds. Pred- dressed to better understand these predator-prey ator confusion is one reason why prey species relationships. evade predatorsby assemblingin large flocks ca- ACKNOWLEDGMENTS pable of cohesivemovements (Curio 1976, Davis 1980). Falcons in this study were nnsuccessfulin I thank Lori Salzer and Mike Finger for accompanying me in the field at beach sites.Lenny Brennan, Anna Ca- all attempts to capture dunlins during stoopsover hall, Mike Finger,TodJohnson, and Terry Schickassisted crestingwaves even though stoopsover calm water with field work at estuarine sites where field work was were commonly made and often successfulat es- partially funded by NSF-SOS Grant SPI80-04760. Keith tuaries (Buchanan et al. 1988). Successfulcaptures Bildstein, Tom Cade, and Jim Watson provided com- ments that improved the manuscript. over water occurred during horizontal pursuit of birds isolatedfrom flocks;it is likely that any con- LITERATURE CITED /hsion effect is negligibleduring horizontal pursuit BAKER,J.A. ANt) R.J. BROOM. 1981. Distribution patterns of a single bird. These resultsappear to support a of raptors in relation to densi• of meadowvoles. Con- predator-confhsion hypothesis. Finally, the lower dor 83:42-47. rate of hunting successat beaches did not appear BEAL,ILG. AN•) H.J. I•mis. 1990. Stalistical analysisof to be related to densityof' potential kleptoparasites, a problem data set: correlated observations. Cortdor which were more abundant at estuarine sites. 92:248-251. According to the ideal free model of habitat se- BECHARD,M.J. 1982. Effect of vegetative cover on fox- lection (Fretwell and Lucas 1970), both dunlins aging site selection by Swainson's hawk. Condor84' 153-159. and falconsshould attempt to spend as much time BERTOCHI,L.E., G. G&STRO^N•)J.P. MYERS.1984. Notes as possiblein habitats that serve to maximize in- on the predators, especiallythe peregrine, of sand- elusive fitness. This means that shorebirds should erlings on the Peruvian coast. WaderStudy Group Bull forage in habitatswhere invertebrateprey is avail- 42:31-32. able at a level that offsetsphysiological costs asso- B•Lt)STEm,K.L. 1987. Behavioral ecology of red-tailed ciated with roosting, vigilance, and predator avoid- hawks (Buteojamaicensis), rough-legged hawks (Buteo ance. Similarly, merlins and peregrine falcons lagopus), northern harriers (Circus cyaneus), and 98 BUCHANAN VOL. 30, NO. 2

American keNtreiN(Palco sparverius) in south central tion in birds. I. Theoretical development.Acta Biotheor. Ohio. Ohio Biol. SurveyBiol. Notes 18. 19:16-36. BOYCE,D.A., Jp,. 1985. Merlins and the behavior of win- HEJI., S.J.,j. VERNE}•^Nr> G.W. BEI.I.. 1990. Sequential tering shorebirdsßRaptor Res. 19:94-96. versusinitial observationsin studiesof avianforaging. BRENNAN,L.A., J.B. BUCHANAN,C.T. SCHICK,S.G. HERMAN Stud. Arian Biol. 13:166-173. ANDT.M. JOHNSON.1984. Sex determination ofdun- HUNT,W.G., R.R. ROGEP,S ANt> D.J. SI•O•N.1975. Migra- lins in winter plumage.J. Field Ornithol.55:343-348. tory and foraging behavior of peregrine falcons on , J.B. BuCI•NAN, S.G. HEP,MAN AND T.M. JOHNSON. the Texas coast. Can. Field-Nat. 89:111-123. 1985. Interhabitat movements of wintering dunlins in Kus, B.E. 1985. Aspectsof flocking behavior and pred- westernWashington. Murrelet 66:11-16. ator avoidance in wintering shorebirds. Ph.D. disser- BROCKMANN,H.J. ANDC.J. BARNARD.1979. Kleptopara- tation, Univ. California, Davis, CA U.S.A. sitism in birds. Anim. Behav. 27:487-514. LEGER, D.W. AND I.A. DIDRICHSONS. 1994. An assessment BUCHANAN,J.B. 1988. The effect of kleptoparasiticpres- of data pooling and some alternatives.Anim. Behar. 48:823-832. sure on hunting behavior and performance of host MOP,PriSON, M.L. 1984. Influence of sample size and sam- merlins.J. RaptorRes. 22:63-64. ß 1992. Winter abundance of shorebirds at coastal pling design on analysisof avian foraging behavior. Condor 86:146-150. beachesof Washington. WashingtonBirds 2:12-19. PAGE,G.W. ANDD.F. WHITACRE.1975. Raptorpredation , S.G. HEP,MAN •_ND T.M. JOHNSONß1986. Success on winteringshorebirds. Condor 77:73-83. ratesof the peregrinefalcon (Palcoperegrinus) hunting PAUI.SON, D.R. 1993. Shorebirds of the Pacific North- dunlin (Calidrisalpina) during winter. RaptorRes. 20: west. Univ. Washington Press,Seattle, WA U.S.A. 130-131ß PRESTON,C.R. 1990. Distribution of raptor foraging in , C.T. SCHICK, L.A. BRENNAN AND S.G. HERMAN. relation to prey biomassand habitat structure. Condor 1988. Merlin predation on wintering dunlins: hunt- 92:107-112. ing successand dunlin escapetactics. Wilson Bull. 100: RATCLIFFE,D. 1980. The peregrine falcon. Buteo Books, 108-118. Vermillion, SD U.S.A. --, C.T. SCHmK, L.A. BRENNAN AND S.G. HERMAN. RO•LKVAM,R. 1985. How effective are hunting pere- 1991. Recoveryof prey from water by merlins.j. Rap- grines?Raptor Res. 19:27-29. tor Res. 25:43-44. SODHI,N.S., t.W. OLIPHANT,P.C. JAMES AND I.G. WARKEN- BUP,GEP,, J. 1988. Effectsof age on foraging in birds. Acta TIN. 1993. Merlin (Falcocolumbarius). Pages 1-19 ,n Cong,:Int. Ornithol.19:1127-1140. A. Poole and F.B. Gill [EDs.], The birds of North Cu•,IO,E. 1976. The ethologyof predation.Spinger Ver- America. Academyof Nat. Sci., Philadelphia,PA and lag, New York, NY U.S.A. Am. Ornithol. Union, Washington,DC U.S.A. DAvis,J.M. 1980. The coordinated aerobaticsof dunlin SWENSON,J.E. 1979. The relationshipbetween prey spe- flocks. Anita. Behar. 28:668-673. cies ecology and dive successin ospreys.Auk 96:408- DE•mEP,,D. 1980. Hunting successrates, foraging habits, 412. and prey selection of peregrine falcons migrating TOLAND,B. 1987. The effect of vegetativecover on for- throughcentral Alberta. Can.Field-Nat. 94:371-382. aging strategies,hunting successand nesting distri- DOBLEP,,F.C. AND R.D. SPENCEP,.1989. Wintering pere- bution of American keNtreiNin central Missouri. J. grine falcon halcoperegrinus habitat utilization in Grays RaptorRes. 21:14-20. Harbor, Washington.Pages 71-78 in B.-U. Meyburg TRELEAVEN,R. 1980. High and low intensityhunting in raptors. Zeitschriftfiir •l•erpsycholo•e54:339-345. and R.D. Chancellor [EDs.], Raptors in the modern WAKEI.EY, J.S. 1978a. Factorsaffecting the useof hunting world. World Working Group on Birds of Prey and sitesby fbrruginoushawks. Condor 80:316-326. Owls, Berlin, Germany. 1978b. Hunting methodsand factorsaffecting EVENSON,J.R. ANDJ.B. BucI4•'qAN. 1995. Winter shore- their useby ferruginoushawks. Condor 80:327-333. bird abundance at Greater Puget Sound estuaries:re- 1979. Use of hunting methods by ferruginous cent census results and identification of potential hawksin relation to vegetationdensity. Raptor Res. 13. monitoring sites.Pages 647-654 in Puget Sound Re- 116-119. search '95. Puget Sound Water Quality Authority, Z•a•, J.H. 1984. Biostatisticalanalysis. 2nd ed. Prentice Olympia, WA U.S_A. Hall, EnglewoodCliffs, NJ U.S.A. FRETWELL,S.D. AND H.L. LUC•S. 1970. On territorial be- havior and other factorsinfluencing habitat distribu- Received25 September1995; accepted2 February 1996