THE INFLUENCE OF TIDE AND WEATHER ON ...

10 downloads 46 Views 1MB Size Report
Mar 3, 2005 - encing provisioning rates at Bald Eagle nests along ... (2002) for Wisconsin. ..... eagle pairs in the Columbia River estuary were "special-.
THE JOURNAL OF RAPTOR RESEARCH A QUARTERLY PUBLICATION

OF THE

VOL. 39

RAPTOR RESEARCH FOUNDATION,

MARCH 2005

INC. NO.

j. RaptorRes.39(1):1-10 ¸ 2005 The Raptor ResearchFoundation, Inc.

THE INFLUENCE OF TIDE RATES OF CHICK-REARING

ISLAND,

AND WEATHER ON PROVISIONING BALD EAGLES IN VANCOUVER

BRITISH

KY•E

HAMISH

COLUMBIA ELLIOTT

CanadianWildlifeService,5421 Robertson Road,Delta, BritishColumbiaV4K 3N2 Canada CHRISTOPHER

E. GILL

SimonFraserUniversity,8888 University Drive, Burnaby,BritishColumbiaV5A 1S6 Canada

JOHNE. ELLIOTT 1 CanadianWildlifeService,5421 Robertsort Road,Delta, BritishColumbiaV4K 3N2 Canada ABSTRACT.--We investigatedhow tide and weather (temperature and rainfall) influenced Bald Eagle (Haliaeetusleucocephalus) provisioningrates on the east and west coastsof Vancouver Island, British Columbia during the 1996 breeding season.Eaglesnesting on the west coasthad lessfrequent preydelivery rates (3.2 + 1.0 deliveries/d, N = 6 nests) than those nesting on the east coast (7.2 _+2.8 deliveries/d,N = 7 nests).Prey-deliveryrateswere negativelycorrelatedwith precipitationon the west coast,but not on the east coast.On both coasts,prey-deliveryrateswere negativelyrelated with temperature (r = -0.45), as nestlingshave lower energetic needs during warm weather, and positively related with adult attendance (r = 0.35). Prey-deliveryrates were highest at intermediate tide height and early in the day, reflecting food availability.Intertidal fish were a major component of eagle diet on the eastcoast(plainfinmidshipman[Porichthys notatus]= 46% by frequency)while pelagicfishwere the major componenton the westcoast (Pacificmackerel [Scomberjaponicus] = 92%). We concluded that landscapefeatures (e.g., tidal flats) and weather (e.g., rain and temperature) interact to influence provisioningratesof Bald Eagleson VancouverIsland, and that thesefhctorsmay drive spatialvariation in eagle productivity.

KEYWORDS: BaldEagle;Haliaeetusleucocephalus;BritishColumbia; ptvductivity; weather;, tides.

LA INFLUENCIA

DE LA MAREA

Y EL CLIMA

EN

LAS TASAS

DE APROVISIONAMIENTO

DE HAL-

IAEETUS IEUCOCEPHALUS QUE SE ENCONTRABAN CRIANDO EN VANCOUVER ISLAND, BRITISH COLUMBIA

P•SUmEN.--Investigamos c6mo la marea y el clima (temperatara y precipitaciones)influenciaron las tasasde aprovisionamientopor parte de individuosde Haliaeetusleucocephalus en las costasestey oeste de Vancouver Island, British Columbia, durante la estaci6n reproductiva de 1996. Las figuilasque estaban nidificando en la costaoeste presentaron tasasde entrega de presasmenos frecuentes (3.2 + 1.0 entregas/d, N = 6 nidos) que aquellasque estabannidificandoen la costaeste (7.2 -+ 2.8 entregas/d, N = 7 nidos). Lastasasde entregade presasestuvieronnegativamentecorrelacionadas con la precipitaci6n en la costa oeste, pero no en la costa este. En ambas costas,las tasasde entrega de presas estuvieronrelacionadasnegativamentecon la temperatura (r = -0.45), debido a que los pichones

• Email address: [email protected]

1

2

ELLIOTT ET AL.

VOL. 39, NO. 1

tienen necesidadesenergaticasmenores durante periodosc/tlidos,y relacionadaspositivamentecon la presencia de los adultos (r = 0.35). Las tasasde entrega de las presasfueron m/rsaltas a nivelesintermediosde altura de la mareay al comienzodel dia, reflejando la disponibilidadde alimentos.Lospeces de la zona intermareal constituyeronun componente importante de la dieta de las •tguilasen la costa este (Pa•ychthys notatus= 46% por frecuencia),mientrasque los pecespelagicosconstituyeronel componerite m/rsimportante en la costaoeste (Scomberjaponicus = 92%). Concluimosque lascaracteristicas del paisaje(e.g., planosmareales)y el clima (e.g., 11uvia y temperatura)interactfianpara influenciarlas tasasde aprovisionamientoen H. leucocephalus en Vancouver Island, y que estosfactores podrian conducir a variacionesespacialesen la productividad de las/tguilas. [Traducci6n del equipo editorial]

Food availabilityis a key factor regulatingraptor (Newton 1978, 1998, Kruger and Lindstrom2001, productivityin natural ecosystems (Newton 1979, McDonald et al. 2004). For example, rain reduced 1998), including Bald Eagles (Haliaeetusleucoce- the ability of Ospreysto locate prey by increasing phalus;Hansen 1987, Dykstraet al. 1998), White- the number of ripples on the water surface (Grubb tailed Sea Eagles (Haliaeetus albicilla; Helander 1977). In addition, the number of rainy daysdur1985), African Fish-Eagles(Haliaeetusvocifer;, Harp- ing early springwasnegativelycorrelatedwith Bald er et al. 2002), Wedge-tailedEagles(Aquilaaudax; Eagle productivityin southeasternAlaska (Gende Ridpath and Brooker 1986), Golden Eagles(Aquila et al. 1997), but not in the drier interior regions chrysaetos; Watson et al. 1992), and Ospreys (Pan- of Alaska (Steidl et al. 1997). &on haliaetus; van Daele and van Daele 1982), In this study, we investigated the factors influamongothers (Platt 1976,Dykstraet al. 2003, Hak- encing provisioningratesat Bald Eagle nestsalong karainen et al. 2003). Although some Bald Eagle the Vancouver Island shoreline in coastal British populations are still at least partly influenced by Columbia. Elliott et al. (1998) and Gill and Elliott toxic contamination (Anthony et al. 1993, 1994, (2003) showed that eagles nesting on the west 1999, Kumar et al. 2002, Bowerman et al. 2003, coasthad lower productivity than those nestingon Dominguez et al. 2003), several populations now the east coast,and conjecturedthat this was due appear to be regulatedby natural agents,primarily to wetter springsand fewer tidal flats limiting the food abundance (e.g., Dzus and Gerrard 1993, amount of food brought to nests.We therefore hyDonaldson et al. 1999, Anthony 2001, Stout and pothesized that (1) provisioning rates would be Trust 2002, Gill and Elliott 2003). negativelycorrelated with tide height and (2) proLandscape features, especiallytidal flats, influ- visioning rates would be lower during periods of ence eagle-preyavailability,and therefore, produc- rain. tivity (Hansen 1987, Watsonet al. 1991, Dzus and Previousinvestigators(Gerrard et al. 1979, BorGerrard 1993, Gende et al. 1997, Wilson and Armtolotti 1984a, Jenkins 1989, Kennedy and Mcstrong 1998, Watson 2002). Weather also influenc- Taggart-Cowan1998, Warnke et al. 2002) have dees raptor reproductive success,either by affecting scribed Bald Eagle nesting behavior using direct adult or chick energetic needs (Newton 1978, observation and time-lapsephotography.Warnke 1998, Stalmaster and Gessaman1984), prey avail- et al. (2002) established several benchmarks for ability (Seavy et al. 1998, Panasci and Whitacre the nest provisioningand chick behavior of an in2000, Harper et al. 2002, Jaksic et al. 1997), or land populationthat did not seemto be limited by adult foragingeffectiveness (Grubb 1977, Newton food. In this paper,we used both methodsto com1978, Stinson 1980, Gilchrist and Gaston 1997, pare the behaviorof eaglesfrom coastalVancouver Gilchrist et al. 1998, Fritz 1998). Whereas rainfall Island with those presented by Warnke et al. usuallyincreasesforaging successfor tropical rap- (2002) for Wisconsin. tors, due to increasedproductivityat lower trophic STUDY AREA AND METHODS levels during the rainy seasonand decreasedheat StudyArea. During May-July 1996, we monitored Bald stressin canopy nesters (Seavyet al. 1998, Panasci and Whitacre 2000, Harper et al. 2002,Jaksicet al. Eagle nestson the east (Crofton, 48.9øN,123.7øW)and west (Barkley Sound,48.9øN,125.3øW)coastof southern 1997, Touchton et al. 2002), rainfall usuallyreduc- Vancouver Island, British Columbia. On the west coast, es prey availability,increaseschick energetic needs, the marine environment drops off steeply from shore and adult-nest attendance in temperate regions while the shelteredwaterson the eastcoastinclude large

MARCH 2005

BALD EAGLE PROVISIONING

tidal flats where many eaglesforage. In addition, the exposed west coast tends to be cooler and windier, with more storm events, than the protected east coast, and nestingbeginsca. 1 mo later (Elliott et al. 1998). Within each study area, we selectednestswhere placement of video recording equipmentafforded a good vantagefor nestingobservations. Observations.We positioned six (three on each coast) miniature CCD cameras (V-1205, Sony,Vancouver,BC, Canada) 2 m above the nestswhen nestlingswere ca. 4 wk old. Our observationscorrespondedca. to the "middle nesting stage" identified by Warnke et al. (2002). Cameras were housed in waterproof compartments, camouflaged using paint and vegetationand connectedvia power and RG-8UM video lines (RG-8UM, CB World, Lansing,MI U.S.A.) to a DC VCR (PanasonicAG-1070 one frame.s •, Secaucus,NJ U.S.A.) on the ground. We replaced the videotapesevery 3 d. VCRs were housed in waterproof compartments 20 m from the nest tree and programmed to record from 0500-1100 H and 14302030 H (812 hr total footage). Two 12-volt deep-cyclemarine batteries in protective housing supplied power. We analyzedvideo recordingsby viewingon a television monitor.

We conducted direct observationsusing 20-60X spotting scopesfrom a blind at four nestson the east coast and three on the west coast (-->6 d/nest, each 16 contin-

uous hours dawn to dusk) for 1344 hr total. Nestlings were 2-10 wk old. Observersworked in pairs 100-300 m from the nest, with individual observersswitchingevery 2 hr (Warnke et al. 2002).

To corroborate the compatibilityof observationmethods, we collected direct observation and video data si-

multaneouslyfor 20 hr. As was found by Dykstra et al. (1998) and Warnke et al. (2002), both methods had

good compatibility(>98% agreement) for determining the number of prey deliveries,prey speciesdelivered,and chick and adult behavior.We consequentlypooled both data sets.

We recorded adult nest attendance, number and type (classand specieswhen possible)of prey delivered, and

eaglet behavior (standing, feeding, preening, fighting with siblings,exercisingwings,playingwith nestmaterial, and walking around the nest; Warnke et al. 2002). All behaviorsexcept "standing" were considered "active." We classifiedprey items into four length categories(715 cm, 15-23 cm, 23-30 cm, 30-40 cm) using the prey item's size relative to adult bill or toe-pad length. Prey length and specieswas then used to estimate biomass using equationsand experimental valuesfrom the literature as describedby Dykstra et al. (1998) and Gill and Elliott (2003). Our analysesfocusedon the relationships betweenvariablesand prey deliveries(rather than energy delivered) as prey-delivery rates were estimated with greater certainty.Environment Canada weather stations at Ucluelet (west coast) and Crofton (eastcoast) provided weather

data.

StatisticalAnalyses.Statisticalanalyseswere performed usingSTATISTICA. Data for individualnestswere pooled on each coast if an analysisof covariance (ANCOVA) gave no significantdifference among nests.We pooled the data for both coasts if the interaction term (coast X

variable) was not significant.We used a multiple linear

RATES

3

regressionto examine the eftkcts of adult nest attendance (percent of each day when at least one adult was present, averagedover each week), mean daily temperature, and chick age (independent variables) on mean daily prey and energydeliveries,averagedover eachweek (dependent variables). Within this multiple linear regression,we used ANCOVAs to compare prey deliveries on dayswith and without rain on each coastand to compare prey deliveries relative to temperature on each coast.In addition, we employed multiple linear regression, using a quadratic model, to examine the effect of tide height, time of day, and prey deliveries (hourly mean) between coasts.Prior to using parametric stat•stics, we tested for homogeneity of variance (Levene's test) and normality (Kolmogorov-Smirnov).Bonferronl adjustmentswere completed on multiple comparisons We consideredresultsto be significantif P < 0.05. The valuesreported are means +_SD. RESULTS

All nestlings present at the beginning of the studyperiod survivedto the end of the studyperiod, with more young produced at the sampled nestson the eastcoast(2 young per occupiednest) than on the west coast (0• = 1.2) of Vancouver Island (Table 1).

Factors Influencing ProvisioningRates. Prey-delivery ratesvaried significantlybetweencoasts.The west coasthad fewer prey deliveries (t = 2.3, df = 8, P = 0.008) per day and lower mean prey biomass (t = 1.9, df = 8, P = 0.04) than the east coast (Table 1). However, there was no difference (t =

2.2, df = 10, P = 0.43) in the mean daily energy delivered (Table 1). There was also no difference

in the prey-sizedistributions(t = 0.73, df = 3, P = 0.25) for Pacific herring (Clupeaharengus),the only speciesfor which there wassignificantoverlap between the two coasts(Table 2). A multiple linear regression(R•2= 0.22, F = 2.61, df = 2103, P = 0.07) indicated no correlation between nestling age and the number of prey deliveries (t = 1.34, df = 202, P = 0.66). However, there was a peak in prey deliveriesat about 4 wk, correspondingto the period of maximum growth (Warnke et al. 2002). Using this regressionwe found a negativerelationship between mean daily temperature and number of prey deliveries(t = -2.35, df = 202, P = 0.02), and the mean number of prey deliveriesand the presenceof rain on the west coast (t = -3.6, df = 103, P < 0.001), but not on the east coast (t = 0.65, df = 98, P = 0.54). In addition, prey-delivery ratesat both coastswere stronglyinfluenced by the time of day (Fig. 1). On both coasts,there was a significant quadratic relationship between tide height and prey-delivery rates (Fig. 2), with the

4

ELLIOTT ET AL.

VOL. 39, No. 1

Table 1. Provisioningratesto VancouverIsland Bald Eaglenestsduring 1996.Valuesindicatedare dailymeans(SD) averagedover the entire nestlingperiod. ENERGY DELIVERED

NESTID

No. NESTLINGS OBSERVATIONS PREY-DELIVERY RATE PERNESTLING(kJ) 2 2 2 2 3 2 I 2.0 I I 2 I 1

6 6 6 14 12 11 11

BS-Sp b

I

11

2.8 (1.5)

2201 (674)

Mean, west

1.2

3.2 (1.0)

2358 (669)

347 (92)

Mean, pooled

1.6

5.4 (2.9)

2147 (880)

451 (224)

• %deo

cameras.

b D•rect

observations.

12 6 6 5 13

greatest prey-deliveryrates at intermediate tide heights. Therefore, we reject the hypothesisthat provisioningrates were negativelyrelated to tide height, at either coasts,and accept the hypothesis that provisioningrates were negativelycorrelated with rain only on the west coast. Adult and Nestling Behavior. On both coasts, adult nest attendancewaspositivelycorrelatedwith brood size (r = 0.33, P < 0.01) and mean number

of prey deliveriesper hour per eaglet (r = 0.34, P < 0.002; Fig. 3), and negativelycorrelatedwith eaglet age (r = 0.27, P < 0.01). Mean daily eaglet activitylevelswere higher on the westcoast (42 +5% of the day) than eastcoast(27 +- 3%) of VancouverIsland. Eagletson the westcoastspent more time standing,fkeding,preening,fighting with sibhngs, exercisingwings,playing with nest material, and walking around the nest than their counterparts on the east coast. On both coasts,eaglet mean daily activity was independent of hatching order (P > 0.6), negativelycorrelated with adult nest attendance (r = 0.46, P < 0.001) and positive-

7.7 10.8 9.0 9.6 4.5 5.0 3.6 7.2 2.3 4.8 4.0 2.6 2.9

(3.5) (3.4) (4.1) (6.6) (2.2) (3.5) (2.3) (2.8) (1.1) (3.2) (1.9) (1.1) (1.4)

1517 3681 3210 872 1490 1539 1448 1965 2314 3547 1691 2584 1811

(345) (780) (2008) (692) (990) (794) (539) (1046) (777) (1060) (1124) (834) (548)

PREYBIOMASS (g)

CN-Bra CN-Cha CN-Mo • CN-WN b CN-WSb CS-MB• CS-PMb Mean, east BS-AIb BS-Cha BS-Nua BS-Ri• BS-SB•

630 955 821 209 380 417 373 541 345 510 236 365 295

(87) (199) (523) (136) (367) (212) (135) (270) (101) (151) (158) (111) (81)

330 (94)

common prey item was plainfin midshipman (Porychthysnotatus)while on the west coastit was Pacific mackerel (Scomber japonicus).At both locations, Pacific herring (Clupea harengus)was the second most common prey item. Direct observations tend to overrepresenteasilyidentified, common speciesso that actual percentagesfor major prey items may be exaggerated(Mersmann et al. 1992, Dykstra et al. 1998). DISCUSSION

Bald Eagle provisioningrateswere clearlyinfluenced by time of day,tide, and weather.High provisioning rates tended to be at intermediate tide heights, at low temperatures,during rainlessperiods, and early during the day. The decline in provisioningrate with time of day likely reflectsboth nestling satiation and declining prey activity,as Watsonet al. (1991) reported in the ColumbiaRiver estuary.

The mean prey-deliveryrate for VancouverIsland eagles(5.4 prey/d) wasremarkablysimilarto the benchmarkof 5.2 prey/d reported by Warnke et al. (2002), which they consideredto indicate adequate prey availabilityto supporthigh Bald Eagle reproductivesuccess. Our data supportthisbench-

ly correlatedwith eaglet age (r = 0.34, P < 0.005; hg. 4). Prey Type. There was considerablevariation in prey type between the east and west coasts.Fish made up the majorityof the diet at both locations mark, as territories on the east coast had consis(Table 2). However, on the east coast the most tently high productivity,while those on the west

MARCH 2005

BALD EAGLE PROVISIONING

RATES

5

Table 2. Prey delivered to VancouverIsland Bald Eagle nestsin 1996. PREYTYPE

PERCENT OF DIETa

NO. OFDELIV-ERIES ENERGETIC VALUE(kJ)b

East Coast

Fish

85.1

215

438

43.0

65

403

10.0 10.9 5.0

23

265

2

3304

I

3075

2

526

Plainfin midshipman (Porichthys notatus)

Pacific Herring (Clupeaharengus) Coho Salmon (Oncorhynchus kisutch) Pollock (Theragrachalcogramma) Flounder (Bothidae/Pleuronectidae

spp.) Ling cod (Ophiodonelongates) Surf Perch (Hyperprosopon ellipticurn)

1.7 1.7 0.6

3

349

I

352

4.0

2

222O

1.7 5.6

i

1046

I

3394

10.9

4

3029

19.6 0.1

2

5967

I

90

Bird

Mallard duckling (Anasplatyrhynchos) Pigeon Guillemot (Cepphuscolumba) Mammal

European rabbit (Sylvilagusspp.) Rodent (Rodentia spp.) West

Coast

Fish

PacificMackerel (Scomberjaponicus) PacificHerring (Clupeaharengus) Coho Salmon (Oncorhynchus kisutch)

94.7

85

963

90.8 6.2 1.9

37

1713

Flounder (Bothidae/Pleuronectidae spp.)

Sculpin (Cottidae spp.) Shiner Perch (Cymatogaster aggregate) Surf Perch ( Hyperprosopon ellipticurn) Bird

Mallard duckling (Anasplatyrhynchos)

14

307

i

1343

1.0 0.1 O.1 0.0

2

334

i

88

1

37

6.5

i

4555

1.2

i

1046

1

15

a Byenergy.

Averagedover all delNeries.

0.3. 0.25

ß

ß

ß ß

0.2

0.05

ß

o.•5

0.1

004

0.05 0

0.5

o.Oo 500

700

I

1.5

2

2.5

3

35

Tide Height (m) 900

1100

1300

1500

1700

1900

Figure 2. Quadratic regressionof provisioningratesfor Bald Eagleson the east (0) and west (m) coastsof VanFigure 1. Linear regressionof mean number of prey couver Island during 1996 relative to tide height (east: deliveriesper hour relative to time of day at Vancouver prey deliveries= -0.75 X [tide height]9 + 0.25 X [tide Island Bald Eagle nestsin 1996 (prey deliveries= -0.033 height] - 0.016, fl = 0.69;west:preydeliveries= -0.052 [time of day] + 0.0458;fl = 0.50). Data were pooledfor X [tide height]a + 0.16 X [tide height] + 0.027; • = Time of day

east and west coasts.

0.68).

6

ELLIOTt ET AL. 100

05 04

VOL. 39, NO. 1 ß

._• 80

•03

ß

60

--

&02

ß

•_ 40

01

• 0

20

40 Adult

60

ß ß

20

8O

Nest Attendence

0

2

4

6

8

10

Nestling Age (wk)

Figure 3. Vancouver Island Bald Eagle adult nest attendance (percent of total observationtime when adult was at nest) relativeto prey-deliveryrate (prey deliveriesper chick per hr). Each data point representsone nest, averaged over the 1996 season(prey deliveries = 0.0016 [adult nest attendance] + 0.22; •e = 0.12).

Figure4. Mean Bald Eaglenestlingactivity(&--percent of total observation time when nestling was active) and adult attendance(I--percent of total observationtime when adult waspresent) relative to nestlingage (nestling activity= 6.1 X age - 3.5; •e = 0.92; adult attendance= -11.3 X age + 97; •e = 0.87). Nests (N = 13) were studied on Vancouver

coast, a region with consistentlylow productivity, were belowthis benchmark.Despitethe difference in number of prey items and biomassdelivered, the total energydeliveredper chick washigher on the west coast than the east coast, although the mean total energy delivered to the nest wassimilar, as east coastnestsaveragedmore chicks.Nestlings on the west coastmet the field energetic requirementsof 2427 ñ 100 kJ/d for nesfiingsin northern Wisconsin (Dykstra et al. 2001a), while those on the east coastdid not. Nestlingsmay have experienced higher energeticdemandson the westcoast due to substantiallycooler conditionsmore similar to Wisconsin.Eagleson the west coastforaged primarily on Pacific mackerel, a high-energy species (1713 kJ;Ann 1973) that wasunusuallyabundant due to surface water warming in the late 1990s (Gill and Elliott 2003). Thus, eagleson the west coastseemto make up for the low quantity of prey deliveredby increasingprey quality (Wright et al. 1998, Hilton et al. 1998), as has been known to

Island

in 1996.

availabilitycan affect spatialvariationin productivity. The east-coastpopulation, which consistently produced more young than needed to replace individualslost to mortality (Elliott et al. 1998), had higher prey-deliveryrates than the west-coast population, which did not produce enough young to compensatefor mortality (Elliott et al. 1998). The relationship between tide height and provisioningrates highlights the importance of tidal flats for nesting eagles.The influence of tide on Bald Eagle foraging habits has been documented in Alaska (Ofelt 1975, Wilson and Armstrong 1998), British Columbia (Hancock 1964, Elliott et

al. 2003), and Washington (Watson et al. 1991, Garrett et al. 1993), and appearsto be characteristic of coastalpopulations in the Pacific Northwest. The peak in prey deliveriesat intermediate tide heightsis likely becausethe plainfin-midshipman-breedingzone is exposedat this tide level (Elliott et al. 2003); in the Columbia River estuary, where eaglesforage primarily on fish that are most easilycaptured in shallowwater, foraging success peaks at low tide (Watson et al. 1991). Weather also influenced provisioningrates.The decline in provisioning rates with temperature likely reflects lower chick-energydemandsduring warm weather, while the lower provisioningrates on rainy days may reflect reduced adult foraging efficiencydue to altered visibility(Grubb 1977) or

occurin other raptors(Barton and Houston 1993). Many studieshave showna strong relationship betweenfood abundanceand reproductivesuccess in raptors. For example, reproductive successin many temperate and arctic raptors closelyfollows changesin small mammal populations (e.g., McInvaille and Keith 1974, Smith et al. 1981, Korpimaki 1992, Wiehn and Korpimfki 1997), while fish abundance influences productivity in piscivores such as Ospreys(van Daele and van Daele 1982), prey behavior(Newton 1978). Furthermore,rain White-tailed Sea Eagles (Helander 1985), African increaseseagle energetic requirementsby up to Fish-Eagles(Harper et al. 2002), and Bald Eagles 21% (Stalmaster and Gessaman 1984), and adults (Hansen 1987, Dzus and Gerrard 1993). Our re- may need to eat more during rainy days,leaving suits demonstrateone mechanismby which prey lessprey for them to deliver to their nestlings.The

MARCH 2005

BALD EAGLE PROVISIONINC RATES

7

larger impact of rainfall on the outer coaston prey- studywasfish (Table 2), outlining the importance deliveryratessupportsthe conjecturethat low pro- of usingdirect observations,as analysisof prey reductivityon the outer coastis partiallyattributable mains at nests often leads to the underestimation to harsh and unpredictable weather (Gende et al. of fish prey (Todd et al. 1982, Mersmann et al. 1997,Elliott et al. 1998). Stormsimpactthe repro- 1992). For comparison,the proportion of fish in ductivesuccess of other raptors.For example,win- diets of eagles nesting in coastalregions varied ter severitydelaysreproduction and reducespro- from 42% in Louisiana (Dugoni et al. 1986) to ductivity in Bald Eagles from northern 71% in the Columbiaestuary(Watsonet al. 1991), Saskatchewan (Gerrard et al. 1992), as well as and 76-85% in southeasternAlaska (Ofelt 1975). Golden Eagles (Tjernberg 1983, Steenhof et al. While Watsonet al. (1991) noted that severaleagle 1997) and Gyrfalcons(Pooleand Bromley1988), pairsin the ColumbiaRiver estuarywere "specialwhile hail stormscaused>30% of the reproductive ists," preferentially taking certain food species, failure in Swainson'sHawks in North Dakota (Gil- such aswaterfowl,relative to other eagle pairs,we more and Stewart 1984). found no suchspecialization.Any variation in diet Food availabilitymay alsoimpact nestlingsurviv- breadth could be traced to differences in local al indirectly.Adults with high prey-deliveryrates food availability,suchasthe presenceof large tidal have higher nest attendance,becauseeaglesthat fiats near neststhat consumedlarge numbers of are more successfulhunting can spend less time

plainfin midshipman.

foraging (Watson et al. 1991), reducing eaglet We concludethat time of day,temperature,rainmortality through predation or exposure.This is fall, and tide height all impact eagle provisioning particularly critical during the first 4 wk when in- rates, and therefore, adult nest attendance and creasednest attendanceaids in nestlingthermo- chick activity.Our resultssupport the hypotheses regulation (Warnke et al. 2002). advancedby Elliott et al. (1998) and Gill and ElliBrown (1993), Knight and Knight (1983), Han- ott (2003) that nesting successis greater on the sen (1986), Knight and Skagen (1988), Watsonet east coast because there are more tidal fiats and al. (1991), and Bennetts and McLelland (1997) showedthat the ability of eaglesto obtain food increaseswith age. As with many other bird species, reproductive successin raptors usually increases with age (Newton 1998); fecunditypeaksat age 5 yr among Eurasian Sparrowhawks(Accipiternisus; Newton and Rothery 2002) and at age 7 yr among Northern Goshawks(Accipitergentilis;Nielsen and Drachmann 2003), declining thereafter in both species.In addition adult-adult SpanishImperial Eagle (Aquila adalberti) pairs monopolize high quality territories, hence have higher averagefecundity than juvenile-juvenile pairs (Ferrer and Bisson2003). Therefore, older individualspresumably provisionat higher rates than youngerones. We did not know the age of adultsin our study, and this factor may explain someof the variability between

individual

nests.

fewer storms. Steenhof et al. (1997) documented

similarresultsfor Golden Eaglesin Idaho, in which reproductive successwas determined by prey (black-tailedjackrabbit [Lepuscalifornicus])abundance and weather (frequencyof hot days). ACKNOWLEDGMENTS

C. Hurd, C. Coker, S. Lee, and R. Gill aided during observations.D. Haycock climbed all nest trees w•th graceand style.D. and C. Braggins,the Mouats,R. Hopping, A. andJ. Caldwell,and S. Napier graciouslyallowed accessto their property.D. Henry and G. Biscallprovided valuableinformationon Bald Eagleuse of their fish bycatch. The Canadian Wildlife Service, a Simon Fraser

UniversityGraduate ResearchFellowship,and the Bald Eagle Rescueand ResearchFoundation provided funding. We are especiallygrateful to J. Bednarz,C. Dykstra, M. Hipfner, B. Steidl, K. Warnke, and J. Watson. Their commentsgreatlyimproved this manuscript.A. Fabro located some obscure literature.

Examination of prey remains at nestsfrom both studyareassuggestedthat birdswere the primary LITERATURE CITED food source,as have most prey remainsstudiesin ANN,H.B. 1973. Studieson the age and growthof the the Pacific Northwest (e.g., Norman et al. 1989, jack mackerel ( Trachurusjaponicus).Bull. Fish. Res Vermeer et al. 1989, Vermeer and Morgan 1989, Agency10:73-82. Knight et al. 1990, Watson 2002). However, birds ANTHONY,R.G. 2001. Low productivityof Bald Eagleson made up lessthan 7% of prey at either location Prince of WalesIsland, southeastAlaska.J. RaptorRes 35:1-8. during our study,asrevealedby direct observation. R.W. FRENZEL,F.B. ISAACS,AND M.G. GARRETT The major prey source at both locations in our --,

8

ELLIOTT ET AL.

VOL. 39, No. 1

1994. Probable causesof nestingfailuresin Oregon's

food deliver rates or environmental contamination?J

Bald Eaglepopulation.Wildl.Soc.Bull.22:576-582.

Great Lakes Res. 24:22-34.

1993. Factorsinfluencing , M.J. GARRETT, ANDG.A. SCHUI.ER. 1993. Environ- Dzus, E.H. ANDJ.M. GERRARD. mental contaminantsin Bald Eaglesin the Columbia Bald Eagledensitiesin north-centralSaskatchewan. J. River estuary.J. Wildl.Manag.57:10-19. Wildl.Manag. 57:771-778. , A.K. M•LES,J.A. ESTES, ANDEB. ISAACS. 1999. Pro- ELLIOTT,J.E., I.E. MOUL, ANDK.M. CHENG.1998. Variable ductivity, diets, and environmental contaminantsin reproductivesuccessof Bald Eagleson the BritishConestingBald Eaglesfrom the Aleutian archipelago. lumbia coast.J. Wildl. Manag. 62:518-529. Environ. Toxicol. Chem. 18:2054-2062. ELLiOTt, K.H., C.L. STRUm,ANDJ.E. ELLiOTt. 2003. Bald BARTON,N.W.H. ANDD.C. HOUSTON.1993. A comparison Eagles, Haliaeetusleucocephalus, feeding on plainfin of digestiveefficiencyin birds of prey. Ibis 135:363midshipman, Porichthusnotatus,at Crescent Beach 371.

Can. Field-Nat.

BENNETTS,R.E. AND B.R. MCLELLAND.1997. Influence of

age and prey availabilityon Bald Eagle foragingbehavior at Glacier National Park, Montana. Wilson Bull. 109:393-409.

BORTOLOTTI, G.R. 1984a. Physicalgrowth and development of nestling Bald Eagleswith emphasison the timing of growthevents.WilsonBull. 96:524-542. --. 1984b. Criteria for determining the age and sex

117:601-604.

FERRER, M. ANDI. B•SSON.2003. Age and territory-quality effectsof fecundity in the SpanishImperial Eagle (Aquila adalberti). Auk 120:180-186.

FRITZ,H. 1998. Wind speed as a determinant of kleptoparasitism by Eurasian Kestrel Falco tinnunculuson Short-eared Owl Asiofiammeus.J. Avian Biol. 29:331333.

GARRETT, M.G., J.W. WATSON,ANDR.G. ANTHONY.1993. of nestlingBaldEagles. J. FieldOrnithol. 55:467-481. Bald Eagle home range and habitat usein the ColumBOWERMAN, W.W., D.A. BEST,J.P. GIESY,M.C. SHIELDCASbia Riverestuary.J. Wildl.Manag.57:19-27. TLE, M.W. MEYER,S. POSTUPALSKY, ANDJ.G. S•KARSVaE.GENDE,S.M., M.E W•LSON,AND M. JACOBSEN. 1997. Re2003. Associations betweenregionaldifferencesin poproductive successof Bald Eagles (Haliaeetusleucocelychlorinatedbiphenylsand dichlorodiphenyldichlophalus)and its associationwith habitat or landscape roethylene in blood of nestling Bald Eagles and refeaturesand weather in southeastAlaska.Can.J. Zool. productive productivity.Environ. Toxicol.Chem.22: 75:1595-1604. 371-376.

BROWN,B.T. 1993. Winter foragingecologyof Bald Eagles in Arizona.

Condor 95:132-138.

DOMINGUEZ,L., W.A. MONTEVECCHI,N.M. BURGESS, J. Bp,•Z•L, AND K.A. HOBSON.2003. Reproductive success,environmental contaminants, and trophic status

of Bald Eaglesnesting in easternNeMbundland. J. RaptorRes.37:209-218. DONALDSON, C,.,J.L. SHUTT,ANDP. HUNTER.1999. Organochlorinecontaminationin Bald Eagle eggsand nestlingsfrom the CanadianGreat Lakes.Arch.Environ. Contam. Toxicol. 36:70-80.

DUGONI,J.A., pJ. ZWANK,ANDG.G. FURMAN.1986. Food of nesting Bald Eaglesin Louisiana.RaptorRes.20: 124-125.

--

ANDM.E WILSON.1997. Supplementalfeeding experimentsof nestingBald Eaglesin southeastern Alaska. J. Field Ornithol.68:590-601. GERRARD, J.M., P.N. GERRARD,P.N. GERRaRD,G.R. BORTOLOTTI,ANDE.tl. DZUS.1992. A 24-yearstudyof Bald Eagleson BesnardLake, Saskatchewan. J. RaptorRes 26:159-166.

GERRARD, P.M., S.N. WIEMEYER, ANDJ.M. GERRARD. 1979. Some observationson the behavior of captive Bald Eaglesbefore and during incubation.RaptorRes.13: 57-64.

GILCHRIST,H.G., AJ. GASTON,ANDJ.N.M. SMITH. 1998. Wind and prey nest sites as foraging constraintson an arian predator,the GlaucousGull. Ecology 79:24032414.

DYKSTR&G.R., J.L. I-I•ys, M.M. S•MON,ANDF.B. DANIEL. ANDAJ. {•ASTON. 1997. Effectsof murre nestsite 2003. Behavior and prey of nesting Red-shouldered -characteristicsand wind conditions on predationsby Hawks in southwesternOhio..]. RaptorRes.37:177187.

--,

W.H. IO•sov,

M.W. MEYER, AND D.K. WARNKE.

2001a. Daily energyexpenditureof nestlingBald Eaglesin northern Wisconsin.Condor 103:175-179. --,

M.W. MEYER, K.L. STROMBORG,D.K. WARNKE, W.W. BOWERMAN,AND D.A. BEST. 200lb. Association

GlaucousGulls. Can.J. Zool.75:518-524. GmL,C.E. ANDJ.E. ELLiOTt. 2003. Influence of food supply and contaminantson breedingsuccess of BaldEagles.Ecotoxicology 12:95-112. Gn,MOR•, D.S. ^ND R.E. STEWART.1984. SwainsoH'sHawk

Buteoswainsoni nestingecologyin North DakotaUSA Condor 86:12-18. of low reproductiveratesand high contaminantlevels T.GJ. 1977. Weather-dependentforagingin Osin Bald Eagleson Green Bay,Lake Michigan.J. Great GRUBB, Lakes Res. 27:239-251. preys.Auk 94:146-149.

--,

--,

D.K. W•mNK•, W.H. IOa•sov,

D.E. AN-

DERSEN, W.W. BOWeRMAN, IV, ANDJ.P.G•ESY. 1998. Low reproductiverates of Lake Superior Bald Eagles:low

HAKKARA/NEN,H., S. MVgR& S. KURKI, E. KORPIMAm,A.

NIKUt•, aND V. KOWUNEN.2003. Habitat composition as a determinant of reproductive successof Teng-

MARCH 2005

BALD EAGLE PROViSIONiNG RATES

9

mahn's Owls under fluctuating food conditions. Oihos

Horned

Owl and Red-tailed Hawk in central Alberta,

100:162-171.

Canada.

Can. Field-Nat. 88:1-20.

HANCOCK, D. 1964. Bald Eagleswintering in the southern Gulf Islands, British Columbia. Wilson Bull. 76:117120.

HANSEN,A.J. 1986. Fighting behavior in Bald Eagles: a

testof game theory.Ecology 67:787-797. 1987. Regulation of Bald Eagle reproductive ratesin southeasternAlaska.Ecology 68:1387-1392. HARPER,D.M., M.M. HARPER,M.A. VmAN1, A. SMART,R.B. CHILDRESS,R. ADATIA, I. HENDERSON,AND B. CHEGE.

2002. Population fluctuations and their causesin the African Fish Eagle (Haliaeetusvocifer[Daudin]) at

Lake Naivasha,Kenya.Hydrobiologia 488:171-180. HELANDER, B. 1985. Reproduction of the White-tailed Sea Eagle Haliaeetusalbicillain Sweden. Holarct.Ecol. 8: 211-227.

HILTON, G.M., D.C. HOUSTON, AND R.W. FURNESS.1998.

MERSMANN, TJ., D.A. BUEHLER, J.D. FRASER, ^ND J.K D. SEEGA•.1992. Assessing bias in studiesof Bald Eagle

food habits.J. Wildl.Manag.56:73-78. NEWTON,I. 1978. Feeding and developmentof Sparrow Hawk Accipiternisusnestlings.J. Zool.Lond. 184:465488.

1979. Population ecology of raptors. Buteo Books, Vermillion, SD U.S.A.

1998. Population limitation in birds. Academic Press,San Diego, CA U.S.A. --AND P. ROTHERY.2002. Age-related trends in different aspectsof the breeding performance of individual female EurasianSparrowhawks(Accipiternisus). Auk 119:735-748.

NIELSEN,J.T. ANDJ. DRACHMANN.2003. Age-dependent reproductive performance in Northern GoshawksAccipitergentilis.Ibis 145:1-8.

Which components of diet quality affect retention NORMAN, D., A.M. BRE^ULT, AND I.E. MouI.. 1989. Bald time of digestain seabirds?Funct.Ecol.12:929-939. Eagle incursionsand predation at Great Blue Heron Jaxcsm, F.M., S.I. SILVA,P.L. MESERVE, ANDJ.R.GUTIERREZ.

1997. A long-term study of vertebratepredator responseto an E1Nino (ENSO) disturbancein western South America.

Oihos 78:341-354.

JENFJNS, M.J. 1989. Behaviour of nestling Bald Eagles. Bird Behar. 8:23-31.

KENNEDY, E. AND I. MCTAGGART-COWAN.1998. Sixteen

yearswith a Bald Eagle's,Haliaeetusleucocephalus, nest. Can. Field-Nat. 112:704-706.

colonies. Colon. Waterbirds 12:215-217.

OFELT,C.H. 1975. Food habitsof nestingBald Eaglesin southeast Alaska. Condor 77:337-338.

PANASCI, T. AND D. WHITACP, E. 2000. Diet and foraging behavior of nesting roadside hawksin Peten, Guatemala. Wilson Bull. 122:555-558.

PIATT,J.B. 1976. Gyrfalconnest siteselectionand winter activityin the western Canadian arctic. Can. Field-Nat 90:338-345.

KNIGHT,R.L. AND S.K. SICaGEN. 1988. Agonistic asymmePOOLE,K.G. ANDR.G. BROMLEY. 1988. Natural history of tries and the foraging ecologyof Bald Eagles.Ecology the Gyrfalcon in the central Canadian arctic. Arctic41 69:1188-1194.

--,

31-38.

PJ. RANDOLPH, G.T. ALLEN,L.S. YOUNG,ANDRJ. RIDPATH,M.G. ANDM.G. BROOKER. 1986. The breeding W1GEN.1990. Diets of nesting Bald Eagles,Haliaeetus of the Wedge-tailed Eagle Aquila audax in relation to leucocephalus, in western Washington. Can. Field-Nat. its food supplyin arid westernAustralia.Ibis128:177104:545-551.

KNIGHT, S.K. AND R.L. KNIGHT. 1983. Aspects of food

194.

SEAVY,N.E., M.D. SCHULZE, D.F. WHITACRE, AND M.A. VAS-

finding by winteringBald Eagles.Auh 100:477-484. QUEZ. 1998. Breeding biology and behavior of the KoRP• 'MAKI, E. 1992. Diet colnposition,prey choice,and Plumbeous Kite. Wilson Bull. 110:77-85. breeding successof Long-eared Owls: effects of mul- SMITH, D.G., J.R. MURPHY,AND N.D. WOFFINDEN.1981 tiannual fluctuations in food abundance. Can.J. Zool. Relationships between jackrabbit Lepus califormcus 70:2373-2381. abundance and Ferruginous Hawk ButeoregalisreproKRUGER,O. ANDJ. LINDSTROM.2001. Habitat heterogeduction. Condor 83:52-56. neity affects population growth in goshawkAccipiter STALM•STER, M.V. ANDJ.A. GESSAMAN. 1984. Ecological gentilis. J. Anim.Ecol.70:173-181. energetics and foraging behavior of overwintering KUMAR,K.S., K. KANNAN,J.P. GIESY,AND S. MASUNAGA. Bald Eagles.Ecol.Monog•: 54:407-428. 2002. Distribution and elimination of polychlorinated STEENHOF,K., M.N. KOCHERT,ANDT.L. MCDONALD.1997. dibenzo-p-dioxins,dibenzofurans,biphenyls,and p,p'Interactive effectsof prey and weather on Golden EaDDE in tissuesof Bald Eagles froln the Upper Pengle reproduction.J. Anim. Ecol.66:350-362.

insulaof Michigan. Env. Sci.Tech.36:2789-2796. MCDONALD, P.G., P.D. OLSEN, AND A. COCKBURN.2004.

Weather dictatesreproductivesuccess and survivalin the Australian Brown Falcon Falco berigora. J. Anim. Ecol. 73:683-692.

McINVAILLE, W.B., JR. ANDL.B. KEITH.1974. Predator prey relations and breeding biology of the Great

STEIDL,RJ., K.D. KOZ1E,ANDR.G. ANTHONY.1997. Reproductive successof Bald Eaglesin interior Alaska J. Wildl. Manag. 61:1313-1321. STINSON,C.H. 1980. Weather-dependentforaging success and sibling aggressionin Red-tailedHawks in central Washington.Condor82:76-80. STOUT,J.H. ANDK.A. TRUST.2002. Elemental and organ-

10

ELLIOTT ET AL.

ochlorine residuesin Bald Eaglesfrom Adak Island, Alaska.J. Wildl.Dis'.38:511-517. TJEP, NBERO,M. 1983. Prey abundance and reproductive successof the Golden Eagle Aquila chrysaetos in Swe-

VOL. 39, No. 1

WAKNKE,D.K., D.E. ANDERSEN, CJ.R. DYKSTRA, M.W. MEYER,ANDW.H. KARASOV. 2002. Provisioningrates and time budgetsof adult and nestlingBald Eaglesat in-

land Wisconsinnests.J. RaptorRes'.36:121-127. WATSON, J.W. 2002. Comparative home rangesand food habits of Bald Eagles nesting in four aquatic habitats TOUCHTON,J.M., H.Y. CHENG,ANDA. PALLERONI. 2002. in western Washington. Northwest.Nat. 83:101-108. Foraging ecology of reintroduced captive-bred subadult Harpy Eagles (Havpia havpyja)on Barro Colo, M.G. GARRETT, ANt>R.G. ANTHONY.1991. Foragrado Island,Panama.Ornitol.Neotrop. 13:365-379. ing ecology of Bald Eaglesin the Columbia River esTODD, C.S., L.S. YOUNG,R.B. OWEN,JR., AND F.J. GRAMtuary.J. Wildl. Manag. 55:492-499. LICH. 1982. Food habits of' Bald Eagles in Maine. J. --, S.R. RAE,ANDR. STILLMAN.1992. Nesting density Wildl. Manag. 46:636-645. and breeding successof Golden Eaglesin relation to VAN DAELE,LJ. ANt>H.A. VAN DAELE. 1982. Factors affood supplyin Scotland.J. Anim. Ecol.61:543-550. fecting the productivity of OspreysPandionHaliaetus WIEHN,J. ANDE. KORPIM•KI.1997. Food limitation on brood size: experimental evidence in the Eurasian nestingin westcentral Idaho USA. Condor84:292-299. Kestrel.Ecology 78:2043-2050. VERMEER,K. ANt>K.H. MORGAN.1989. Nesting population, nest sites,and prey remains of Bald Eaglesnest- WILSON, M.F. AND R.H. ARMSTRONG. 1998. Intertidal foring in Barkley Sound, British Columbia. Northwest. aging for Pacific Sand-Lance,Ammodytes hexapterus, by den. Holart. Ecol. 6:17-23.

birds. Can. Field-Nat. 112:715-716.

Nat. 70:21-26.

, --,

R.W. BUTLER, AND G.E.K. SMITH. 1989.

WPdGHT, J., c. BOTH, P.A. COTTON,ANt>D. BRYANT. 1998. Population, nesting habitat, and food of Bald Eagles Quality vs. quantity: energetic and nutritional tradein the Gulf Islands.Pages123-130 in K. Vermeer and offs in parental provisioning strategies.J. Anim. Ecol 67:620-634. R.W. Butler [EDS.],The ecologyand statusof marine and shore birds in the Strait of Georgia, British Columbia. Special Publication, Canadian Wildlife Ser- Received 16 April 2004; accepted 31 October 2004 vice, Ottawa, Ontario, Canada. AssociateEditor: James W. Watson