[219]

The Condor 105:219–227q The Cooper Ornithological Society 2003

MIGRATION ROUTES OF STEPPE EAGLES BETWEEN ASIA ANDAFRICA: A STUDY BY MEANS OF SATELLITE TELEMETRY

BERND-U. MEYBURG1,3, PATRICK PAILLAT2 AND CHRISTIANE MEYBURG1

1World Working Group on Birds of Prey, Wangenheimstr. 32, D-14193 Berlin, Germany2National Wildlife Research Center, P.O. Box 1086, Taif, Saudi Arabia

Abstract. We trapped 16 Steppe Eagles (Aquila nipalensis) on migration and on theirwintering grounds and fitted them with satellite transmitters, 15 of them in Saudi Arabiaand one in South Africa. Seven of the 14 Steppe Eagles trapped in Arabia in autumn didnot migrate to Africa but spent the winter in the Arabian Peninsula. One adult migrated tosouthern Africa. The other six wintered in northeastern Africa, in some cases north of Bab-el-Mandeb, the straits at the southeastern end of the Red Sea, which they had crossed toreach the African continent. On their spring migration all eagles wintering in Africa migratedvia the Suez, Egypt–Eilat, Israel, area at the northern tip of the Red Sea. This loop migrationaround the Red Sea is probably caused by east winds that blow from October until April,making the return migration difficult via Bab-el-Mandeb. This finding should help to explainthe difference in eagle numbers between spring and autumn at such migration bottlenecksas Eilat, Suez, and Bab-el-Mandeb. Unlike eagles coming from Sudan and Ethiopia, eagleswintering in southern Africa must make a considerable detour of over 1200 km to completethis loop. The increase in Steppe Eagles overwintering in Arabia has probably contributedto the decline in the number of birds passing through Eilat in spring during recent years.

Key words: Aquila nipalensis, migration, satellite telemetry, Steppe Eagle, winteringsites.

Rutas Migratorias de Aquila nipalensis entre Asia y Africa: Un Estudio por Medio deTelemetrıa Satelital

Resumen. Atrapamos 16 aguilas (Aquila nipalensis), incluyendo aves migratorias e in-vernantes (15 en Arabia Saudita y una en Sud Africa) y les colocamos transmisores sateli-tales. Siete de las 14 aguilas atrapadas en Arabia durante el otono no migraron hacia Africa,sino que invernaron en la Penınsula Arabica. Aparte de un adulto que migro hacia el surde Africa, las restantes seis invernaron en Africa oriental, en algunos casos al norte de losestrechos al extremo sureste del Mar Rojo (Bab-el-Mandeb) los cuales cruzaron para alcan-zar el continente africano. Durante la migracion primaveral, todas las aguilas que invernaronen Africa migraron vıa Suez y vıa el area de Eilat, alrededor del extremo norte del MarRojo. La migracion alrededor de este cuerpo de agua siguiendo una ruta diferente a lainvernal, es probablemente causada por el viento que provenie del este entre octubre y abril,dificultando la migracion de retorno vıa Bab-el-Mandeb en febrero y marzo. Este descubri-miento debe ayudar a explicar la diferencia en los numeros entre primavera y otono ensitios como Eilat, Suez y Bab-el-Mandeb, que actuan como cuellos de botella. La migracionsiguiendo dos rutas diferentes significa un desvıo considerable de 1200 km para aquellasaguilas que invernan en el sur de Africa, pero no para las que vienen desde Sudan y Etiopıa.El incremento de aves que inviernan en Arabia probablemente ha contribuido durante losultimos anos a la disminucion en los numeros de aves que pasan a traves de Eilat enprimavera.

INTRODUCTION

The Steppe Eagle (Aquila nipalensis) breedsfrom southeastern European Russia to Manchu-ria in eastern China (Meyburg 1994). Next tothe White Stork (Ciconia ciconia), the SteppeEagle, with a weight of 2.4 to 3.9 kg, is the

Manuscript received 22 January 2002; accepted 18November 2002.

3 E-mail: wwgbp@aol.com

largest Palearctic bird species to overwinter reg-ularly and in large numbers in Africa south ofthe equator. During the last 35 years of researchinto raptor migration through the Middle East,the Steppe Eagle has proved an intriguing spe-cies. Although each year well over 100 000 in-dividuals must travel to Africa, very little isknown of the course and dynamics of their jour-ney, or of their behavior on their winteringgrounds (Welch and Welch 1991a, Meyburg1994, Shirihai et al. 2000). Whereas counts have

220 BERND-U. MEYBURG ET AL.

been carried out at certain concentration pointssuch as Suez, Egypt, Eilat, Israel, (summarizedin Bijlsma 1987, Shirihai et al. 2000) and thestraits of Bab-el-Mandeb, at the southern end ofthe Red Sea (Welch and Welch 1988, 1989,1991b), they provide no information on the mi-gration routes. Band recoveries are almost com-pletely lacking (Mikhelson 1982).

In recent years satellite telemetry has provedto be a highly appropriate new method of re-cording the migration routes and dynamics ofindividual birds (Grubb et al. 1994, Meyburg etal. 1995, 2000, Brodeur et al. 1996, Fuller et al.1998, Ueta et al. 1998, 2000, Martell et al. 2001,McGrady et al. 2002). We report here on themovements of 16 Steppe Eagles of differentages and both sexes equipped with satellitetransmitters (PTTs) and tracked by means of theArgos system.

METHODS

TRAPPING TECHNIQUE

From 1993–1998 we captured 15 Steppe Eaglesof varying ages in west-central Saudi Arabia andfitted them with satellite transmitters. Weworked at two rubbish dumps near Taif(218379N, 408439E) and at an overnight roost siteon the beach of the Red Sea near Mecca(20839N, 408259E). One other bird (No. 11) wascaptured in January 1995 in Kruger NationalPark, South Africa, (258039S, 318379E) using abal-chatri trap (Table 1).

In Saudi Arabia we captured the Steppe Ea-gles by vehicle pursuit. The capture team wascomposed of a driver and an onboard observer,who caught the birds. We often used two teamssimultaneously, capturing birds shortly beforeand after sunset (615 min) when thermals weremuch less active. We chased the flying eagleswith one or two four-wheel-drive vehicles. Wesingled out a bird and cut off its flight path threeto five times, at which time it would land on theground. During pursuit, we drove as closely aspossible to the eagle, forcing it to make erraticflight maneuvers, which we could easily match.After the eagle landed, it was crucial to secureit quickly, as birds would fly or run several doz-en meters farther if missed. Once the eagle wasrunning, pursuit by the observer was necessarybut never lasted more than 30 sec. We thenthrew a blanket over the bird and the observerseized it, with gloved hands, by the wings and

legs. We then hooded the bird and allowed it torecover prior to handling.

We recorded the time between initial take-offand final capture (capture time) with a stop-watch. We abandoned chases when birds werenot captured within 15 min because we pre-sumed that longer chases could be detrimentalto them. The capture rate was very high (92%)and demonstrated the efficiency of this method.The mean capture time in our study was lessthan 10 min and was comparable to helicopterpursuits of Golden Eagles (Ellis 1975). We ob-served no mortality during the 18–22 hr follow-ing capture. Three Steppe Eagles that were notfitted with PTTs were recaptured more than oneyear after their first capture using the samemethod. In another study an Imperial Eagle (A.heliaca) was similarly trapped three times in dif-ferent years (Meyburg et al., unpubl. data). Al-though survival of birds not captured by thistechnique is unknown, these data suggest thatcapture stress did not jeopardize postrelease sur-vival.

We measured wing length, tail length, beaklength, wingspan, and weight of each bird andbanded each with a National Commission forWildlife Conservation and Development band(Riyadh, Saudi Arabia). Age was determined byreferring to Clark (1996) and sex by size andweight. All captured birds received a thoroughmedical check by a veterinarian and were eval-uated for degree of cardiovascular stress (Os-trowski et al. 2001).

Because of the dark of night following han-dling, birds could not be released the same day.Also, we kept birds in total isolation in a shadedaviary, even though this was stressful to them,to document any immediate postcapture mortal-ity. For these two reasons, we released the birdsat the pursuit site 18 to 22 hr after capture. Dur-ing capture and handling, we followed the ani-mal welfare guidelines of the National Commis-sion for Wildlife Conservation and Developmentof Saudi Arabia.

This technique permits capture and recaptureof selected individuals and has been useful forother large raptors, such as Imperial Eagle,Greater Spotted Eagle (A. clanga) and Lappet-faced Vulture (Torgos tracheliotus). Althoughvery effective in catching wintering and migrat-ing eagles, efficient use of this technique is lim-ited to open desert and steppe habitats, free ofobstructions.

MIGRATION ROUTES OF STEPPE EAGLES 221

TA

BL

E1.

Det

ails

of16

Ste

ppe

Eag

les

trac

ked

wit

hsa

tell

ite

tran

smit

ters

betw

een

1993

and

1998

inA

sia,

Ara

bia,

and

Afr

ica.

Nor

ther

nmos

tan

dso

uthe

rnm

ost

loca

tion

sar

epr

ovid

edfo

rbi

rds

trac

ked

duri

ngth

ebr

eedi

ngan

dw

inte

ring

seas

ons,

resp

ecti

vely

.A

ge:

juve

nile

sar

e#

1ye

arol

d;ad

ults

are

$4

year

sol

d.P

lace

sof

capt

ure:

Taif

and

Mec

ca,

Sau

diA

rabi

a;K

ruge

rN

atio

nal

Par

k,S

outh

Afr

ica.

Eag

leA

geS

exW

eigh

t(k

g)P

lace

ofca

ptur

eT

rack

ing

date

s

Tra

ckin

gpe

riod

(day

s)N

orth

ernm

ost

loca

tion

Sou

ther

nmos

tlo

cati

on

E1

E2

E3

E4

E5

E6

E7

E8

adul

tju

veni

leim

mat

ure

adul

tad

ult

juve

nile

adul

tju

veni

le

M F F F M F M

2.45

2.75

3.00

2.95

2.25

3.05

2.45

Taif

Mec

caTa

ifTa

ifTa

ifM

ecca

Taif

Mec

ca

2N

ov19

92–1

0M

ar19

9320

Oct

1993

–9D

ec19

9317

Mar

1993

–10

Jun

1993

24O

ct19

93–3

0Ju

n19

9424

Oct

1993

–24

Mar

1994

20O

ct19

93–1

7M

ar19

9424

Oct

1993

–15

Mar

1994

20O

ct19

93–2

6M

ay19

94

129 47 86 319

152

133

145

215

5083

19N

,52

8489

E46

8269

N,

5881

29E

4480

09N

,53

8099

E

2183

59N

,40

8529

E6 8

319N

,40

8059

E21

8249

N,

4085

19E

1880

59N

,44

8319

E24

8079

S,

2782

29E

3 833

9S,

3584

59E

2183

39N

,40

8469

E2 8

109S

,34

8259

EE

9E

10E

11E

12E

13E

14E

15E

16

imm

atur

e(2

year

s)ad

ult

adul

tad

ult

imm

atur

e(1

.5ye

ars)

imm

atur

e(3

year

s)ad

ult

adul

t

F F M F F F F M

2.80

3.30

2.25

3.20

3.25

3.25

3.45

2.30

Mec

caTa

ifK

ruge

rTa

ifTa

ifTa

ifTa

ifTa

if

20O

ct19

93–2

7O

ct19

9422

Oct

1993

–3S

ep19

9402

Jan

1995

–31

Mar

1995

27O

ct19

96–1

1A

ug19

9723

Oct

1996

–23

Apr

1997

26O

ct19

96–8

Dec

1996

27O

ct19

96–7

Mar

1997

21O

ct19

97–1

3D

ec19

98

373

361 89 289

183 44 132

419

5082

79N

,47

8089

E49

8189

N,

6083

59E

4683

99N

,55

8229

E49

8509

N,

6381

69E

4684

49N

,55

8359

E

4783

69N

,58

8519

E

9 814

9N,

2581

89E

1184

79N

,40

8369

E25

8039

S,

3183

79E

2180

89N

,40

8549

E12

8459

N,

4481

39E

2085

09N

,40

8469

E21

8009

N,

4184

69E

1183

89N

,34

8509

E

222 BERND-U. MEYBURG ET AL.

THE TRANSMITTERS

Eleven transmitters were battery powered; thelast five transmitters fitted in 1996 and 1997were solar powered. The weight of the trans-mitters, all supplied by Microwave Telemetry(Columbia, Maryland), was 50–80 g. In order toprolong battery life while adequately recordingthe migration route, these transmitters were pro-grammed to emit signals for 8 hr every 2.3 days(in two cases) and every 4.3 days (in eight cas-es). The transmitter fitted to eagle 11 (E11) wasprogrammed with different cycles to conform tothe eagle’s expected movements (8 hr every dayduring migration, 8 hr every 6 days during win-tering and breeding). The solar-powered trans-mitters fitted to eagles 12–15 were programmedto emit signals for 12 hr every 5.3 days. Thetransmitter fitted to eagle 16 was programmedwith different cycles to conform with the eagle’sexpected movements (to send signals continu-ously during migration and 8 hr every 6 daysduring wintering and breeding). The transmitterswere fitted like backpacks using teflon ribbon toattach them to the birds just prior to release.

DATA ANALYSIS

All location data were analyzed individually andentered into databases. We used the computerprogram Mapit (Allison 1997) to plot Argos lo-cations, measure distances between the loca-tions, and trace the migration routes. This pro-gram is an integrated global mapping and digitaldisplay system which computes the great-circledistance from one point to another while dynam-ically displaying both great-circle and constant-compass-bearing (rhumb) lines. (Great-circledistances are the physically shortest distances ona globe.) Distances between wintering and sum-mering areas and between segments of this werecalculated as the sum of the great-circle distanc-es between all accepted neighboring Argos lo-cations (LC1 5 350–1000 m, LC2 5 150–350m, and LC3 , 150 m location accuracy).

RESULTS

The eagles were tracked for up to 13.7 monthsand for up to 17 100 km, leaving aside localmovements. A total of 3734 locations was usedin this study. From the 11 birds (E1–11)equipped with battery-powered transmitters weobtained 1219 locations, between 18 and 394 pereagle (mean 111 6 108). The length of time be-tween fixing the radio and last location ranged

from 47 to 373 days with an average of 186 6109 days; in some instances the transmitterscontinued to send signals after the birds had ei-ther lost them or perished. From the five birdswith solar-powered PTTs (E12–16) a total of2510 locations was obtained (mean 502 6 834).The time between fitting and last location wasbetween 44 and 419 days (mean 213 6 129days).

Discounting one bird (E3) that was trapped inthe spring and thus presumably had wintered inArabia, seven of the 14 Steppe Eagles trappedin Arabia never left the peninsula for wintering,whilst the other seven crossed the Bab-el-Man-deb straits at the southeastern end of the RedSea to Africa. Of these, six birds remained ineastern and northeastern Africa (Kenya, Tanza-nia, Chad, Sudan, and Ethiopia), thus partly win-tering even north of Bab-el-Mandeb. One adultfemale traveled as far as southern Africa(248079S, 278229E). All eagles wintering in Af-rica conducted a loop migration around the RedSea. Birds on their southward migration movedalong the east coast of the Red Sea; on theirreturn journey they traveled up the west coast ofthe Red Sea (Fig. 1).

Nine eagles which could be tracked to theirsummering areas spent the summer in Russiaand Kazakhstan, just east of Lake Aral(638169E). Two adults, however, possibly bred ineast Kazakhstan or even west Mongolia, as in-dicated by the direction of their northward mi-gration until loss of contact. Only one bird (E9)migrated over the land between the Black andCaspian Seas; the others migrated between theCaspian and Aral Seas. Eagle 10 summerednorth of Lake Aral and performed a loop aroundthis body of water, passing east of it on its south-ward migration (Fig. 1).

Figures 2 and 3, based on analysis of the dataobtained from tracking 16 Steppe Eagles, ourown field observations in Saudi Arabia, Kenya,Tanzania, Zambia, Botswana, Namibia, SouthAfrica, and Israel, as well as data from otherstudies, show, in general terms, our conclusionsregarding the routes followed by these birds onspring and autumn migration between Eurasiaand Africa, excluding individuals wintering inArabia.

DISCUSSION

Satellite telemetry transmitters require consid-erable energy to communicate with satellites,

MIGRATION ROUTES OF STEPPE EAGLES 223

FIGURE 1. Migration routes of four Steppe Eagles(E5, 9, 10 and 16) tracked from Arabia to Africa viathe straits of Bab-el-Mandeb in autumn, and back toArabia and Asia in spring via Suez, Egypt, and Eilat,Israel. All tracked eagles using Bab-el-Mandeb toreach Africa and returning to Arabia and Asia showedthis pattern of loop migration around the Red Sea.

FIGURE 2. General conclusions about Steppe Eagleautumn migration routes between Eurasia and Africa,excluding individuals wintering in Arabia. Conclu-sions are based on our data from 16 satellite-trackedSteppe Eagles and other studies (Christensen and Sor-ensen 1989, Welch and Welch 1991a, Shirihai et al.2000). It remains unclear why a considerable numberof eagles make their way to Africa north of the RedSea via Eilat and Suez (unless there is a migratorydivide in the area indicated by a question mark).

and we found that lifespan of the batteries wasthe fundamental weakness of the system usingbattery-powered transmitters. The number of lo-cations received using solar-powered transmit-ters was substantially higher, so that, particularlywith E16, the route could be plotted far moreaccurately than with transmitters operating onbatteries.

MIGRATION ROUTES AND ORIENTATION

One of the most interesting points to emergefrom this study must surely be the loop migra-tion, described here for the first time, takenround the Red Sea by all the birds that winteredin Africa. This should help to explain the dif-ference in numbers between spring and autumnat such migration bottlenecks as Eilat, Suez, and

Bab-el-Mandeb. Yom-Tov (1984, 1988) dis-cussed this phenomenon at length, maintainingthat following mortality in winter, especially ofyoung birds, the number of eagles on spring mi-gration over Eilat should essentially be smallerthan in autumn. However, the opposite is thecase. He further pointed out that a similar phe-nomenon is to be observed, not only in the caseof other raptor species, in particular Steppe Buz-zard (Buteo buteo vulpinus), Honey Buzzard(Pernis apivorus) and Black Kite (Milvus mi-grans), but also many small bird species.

For the Steppe Eagle population between theAral and Caspian Seas the shortest route to theirwintering grounds in Arabia and Africa runspast the southeast end of the Caspian Sea andnorthwest end of the Arabian Gulf. Joining thesetwo points in a straight line and extending it, one

224 BERND-U. MEYBURG ET AL.

FIGURE 3. General conclusions about Steppe Eaglespring migration routes between Africa and Eurasia,excluding individuals wintering in Arabia. Conclu-sions are based on our data from 16 satellite-trackedSteppe Eagles and other studies (Christensen and Sor-ensen 1989, Welch and Welch 1991a, Shirihai et al.2000).

ends up near Jeddah, on the Red Sea (Fig. 2).This shortest route from the Caspian Sea to theArabian Gulf is apparently taken by manySteppe Eagles, and they do not seem to changedirection very much from there on, so that theyend up far too far north to accomplish the short-est route to Bab-el-Mandeb within the ArabianPeninsula. It remains unclear why a considerablenumber of eagles make their way to Africa northof the Red Sea via Eilat and Suez (24 246 ob-served near Eilat in the autumn of 1980, Shirihaiet al. 2000; 64 900 observed near Suez in theautumn of 1981, Christensen and Sorensen1989, Zalles and Bildstein 2000). Perhaps thereis a migratory divide somewhere south of Eilatthat serves as the deciding factor in route choice.This possibility, and the question of how the dif-ferent migration routes developed, remain to bedetermined.

In this regard the autumn migration of E16was particularly interesting. It was surprising

that E16 reached the Red Sea only 314 km southof Eilat and then flew on southward followingclosely the east coast of the Red Sea. Judgingfrom its route across Arabia one would have ex-pected it to continue over Eilat. In view of thelarge number of Steppe Eagles passing over Ei-lat in autumn (Christensen and Sorensen 1989,Welch and Welch 1991a) other birds arriving ona similar route from the east and northeast mustturn north. This would also explain the frequentarrival during autumn at Eilat of eagles on pas-sage from the southeast (Welch and Welch1991a). Birds migrating through Eilat in autumnneed to be equipped with transmitters there andtracked for at least one year thereafter to eluci-date these questions.

We suppose that for unknown reasons somebirds turn northwest toward Eilat, possibly byfollowing other birds, while others turn south atsome point to arrive at Bab-el-Mandeb. TrackingLesser Spotted Eagles (A. pomarina) over morethan one year showed the development of mi-gration routes of individual birds including as-tonishing detours of hundreds of kilometers(Meyburg et al. 2002). E16 was tracked twicefollowing the Red Sea coast to Bab-el-Mandeband the same wintering area in 1997 and 1998,suggesting that Steppe Eagles may adopt thesame route year after year out of tradition.

For the eagles wintering in southern Africathe loop around the Red Sea involves a consid-erable detour (in the case of E5 this meant anextra 1250 km). For eagles wintering in the Su-dan and Ethiopia (e.g., E10), the loop route doesnot entail a detour. Since Welch and Welch(1991b) by direct observation also found no ev-idence of migrating Steppe Eagles crossing Bab-el-Mandeb in spring we assume that all SteppeEagles entering Africa via this migration bottle-neck follow the loop migration around the RedSea.

The wind blowing from the east during thecool season (October to April) helps the mi-grants coming from the east to cross the straitsin autumn. The average wind speed is 14.5 kmhr21 according to the weather database for thelast 14 years in the Bab-el-Mandeb area (Weat-herbase 2002). This same easterly wind proba-bly prevents them from crossing the straits againwhen returning to their breeding grounds in Feb-ruary and March, thus leading to the evolutionof this migration loop.

MIGRATION ROUTES OF STEPPE EAGLES 225

FLIGHT OVER WATER

According to our direct observations on the RedSea coast near Jeddah, many Steppe Eagles areobliged to roost on the shore in the evening.Many days of such observations led us first toconclude that some eagles elect to cross the RedSea from here. We watched some individualsdisappear from sight across the water. Data fromsatellite telemetry, however, show that on reach-ing this barrier Steppe Eagles change directionand follow the coastline south to the Bab-el-Mandeb Straits.

Whereas high mountains and deserts clearlypresent no insurmountable ecological barriersfor a migrating Steppe Eagle, it, like other largeusers of thermals, avoids crossing large expans-es of water, which can lead to considerable lossof life (Zu-Aretz and Leshem 1983, Kerlinger1985, Spaar and Bruderer 1996). Those eaglesflying to southern Arabia and continuing to Af-rica must, however, cross the 22-km-wide straitsof Bab-el-Mandeb. Locations from all the birdswe tracked showed that in nearly every casethey lingered in Yemen for several days beforeattempting the crossing. We do not know wheth-er they were reluctant to fly over the sea orwhether (more likely) they were waiting formore favorable weather conditions.

The movements of E13 even led to the con-clusion that this bird was prevented from mak-ing the crossing. It first flew in typical fashiondown the Red Sea coast to the vicinity of thestraits. Unfortunately the small number of loca-tions received made it impossible to tell exactlyhow long the eagle stayed there. It can, however,be presumed that for some reason, perhaps dueto weather conditions, E13 was deterred fromcrossing over to Africa. At that time the weatherwas predominantly bad, with heavy rain. Sub-sequent locations indicated that the bird spentthe winter in various parts of the Arabian Pen-insula until 1 April. During this time it wastracked over a total of 5537 km. In early Aprilit left, migrating north through Kuwait and Iran.

After being equipped, E10 stayed from 24 Oc-tober to 15 November relatively close to the RedSea coast before migrating to wintering groundsin the Sudan. Its local back-and-forth move-ments to and from the coast and the fact that itswintering grounds were more or less on the op-posite side of the Red Sea led us to presume thatit was waiting for favorable weather conditions

to make the crossing. Only after three weeks didit fly farther down the coast to make the detourover Bab-el-Mandeb. Similarly, E5 avoided fly-ing over the ca. 45-km-wide Lake Malawi,thereby deviating slightly from the shortestroute.

WINTER DISTRIBUTION

Brooke et al. (1972) advanced the theory onecological grounds that adult and immatureSteppe Eagles have different wintering grounds.This theory has found its way into relevant sum-marizing works (Brown et al. 1982, Steyn 1982,Shirihai et al. 2000), according to which adultbirds overwinter in eastern Africa and youngbirds in southern Africa. The birds we trackedgave this idea little support as did our own fieldobservations in Kenya, Tanzania, Zambia, Na-mibia, and South Africa (Meyburg et al., unpubl.data). Only relatively few individuals flew tosouthern Africa, and these did so independent oftheir age. A greater number spent the winter inEast Africa. In Tanzania we observed the speciesin December and January, in large concentra-tions consisting of adults, juveniles and imma-ture birds.

That a plainly substantial part of the popula-tion overwinters in Arabia was not previouslyknown or has at least not been well documented.It may well be that the use of wintering groundsin Arabia by an increasing part of the SteppeEagle population is a new development resultingfrom an improved supply of food. In recentyears the number of herds of domestic ungulateshas increased dramatically owing to substantialartificial feeding (Shobrak 1996), so that a plen-tiful supply of food is at hand for the eagles. Atsuch places, where the Steppe Eagle is by farthe commonest bird of prey, it is not unusual tosee 50 to 100 birds together. This may have re-sulted in fewer birds migrating as far as easternor southern Africa. The ecological reason forsome individuals to undertake such long migra-tions remains obscure. The possible increase inoverwintering in Arabia has perhaps contributedto the decline in the number of birds passingthrough Eilat in spring (Shirihai et al. 2000) dur-ing the past 10 years.

ACKNOWLEDGMENTS

The authors wish to thank the Deutsche Forschungs-gemeinschaft in Bonn, Germany, for its unstinting fi-nancial support of the Satellite Telemetry Steppe Eagle

226 BERND-U. MEYBURG ET AL.

Project. Wildlife research programs at the NationalWildlife Research Center in Saudi Arabia have beenmade possible through the initiative of His RoyalHighness Prince Saud Al Faisal and under the guid-ance of Professor Abdulaziz H. Abuzinada, SecretaryGeneral of the National Commission for Wildlife Con-servation and Development in Riyadh. We also thankthose who helped with fieldwork, especially Xavier Ei-chaker, Stephane Ostrowski, Philippe Gaucher, and Dr.Alan Kemp. Professor Kai Graszynski read a first draftof the manuscript and was helpful in various otherways. Jean-Yves Cardona provided us with a car andtrapped many eagles for us. Abdulrahman Khojahelped with logistics. We also thank R. D. Chancellorfor linguistic help and Professor Ian Newton as wellas two unknown referees for helpful suggestions.

LITERATURE CITED

ALLISON, J. B. 1997. Mapit. Version 2.0. Allison Soft-ware, Apollo, PA.

BIJLSMA, R. G. 1987. Bottleneck areas for migratorybirds in the Mediterranean region: an assessmentof the problems and recommendations for action.International Council for Bird Preservation, Cam-bridge, UK.

BRODEUR, S., R. DECARIE, D. M. BIRD, AND M. FULLER.1996. Complete migration cycle of Golden Eaglesbreeding in northern Quebec. Condor 98:293–299.

BROOKE, R. K., J. H. GROBLER, M. P. S. IRWIN, AND P.STEYN. 1972. A study of the migratory eagles Aq-uila nipalensis and A. pomarina (Aves: Accipitri-dae) in southern Africa, with comparative noteson other large raptors. Occasional Papers of theNational Museum of Rhodesia B5(2):61–114.

BROWN, L. H., E. K. URBAN, AND K. NEWMAN. 1982.The birds of Africa. Vol. I. Academic Press, Lon-don.

CHRISTENSEN, S., AND U. G. SORENSEN. 1989. A reviewof the migration and wintering of Aquila poma-rina and Aquila nipalensis orientalis, p. 139–150.In B.-U. Meyburg and R. D. Chancellor [EDS.],Raptors in the modern world. World WorkingGroup on Birds of Prey, Berlin.

CLARK, W. S. 1996. Ageing Steppe Eagles. BirdingWorld 9:268–274.

ELLIS, D. H. 1975. First experiments with capturingGolden Eagles by helicopter. Bird-Banding 46:217–219.

FULLER, M. R., W. S. SEEGAR, AND L. S. SCHUECK.1998. Routes and travel rates of migrating Pere-grine Falcons Falco peregrinus and Swainson’sHawks Buteo swainsoni in the Western Hemi-sphere. Journal of Avian Biology 29:433–440.

GRUBB, T. G., W. W. BOWERMAN, AND P. H. HOWEY.1994. Tracking local and seasonal movements ofwintering Bald Eagles Haliaeetus leucocephalusfrom Arizona and Michigan with satellite telem-etry, p. 347–358. In B.-U. Meyburg and R. D.Chancellor [EDS.], Raptor conservation today.World Working Group on Birds of Prey and ThePica Press, Berlin.

KERLINGER, P. 1985. Water-crossing behaviour of rap-tors during migration. Wilson Bulletin 97:109–113.

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 asdetermined by satellite telemetry. Condor 103:715–724.

MCGRADY, M. J., L. T. MAECHTLE, J. J. VARGAS, W. S.SEEGAR, AND M. C. PORRAS PENA. 2002. Migrationand ranging of Peregrine Falcons wintering on theGulf of Mexico coast, Tamaulipas, Mexico. Con-dor 104:39–48.

MEYBURG, B.-U. 1994. 209. Steppe Eagle Aquila ni-palensis, p. 194. In J. del Hoyo, A. Elliott, and J.Sargatal [EDS.], Handbook of the birds of theWorld. Vol. 2. Lynx Edicions, Barcelona.

MEYBURG, B.-U., J. M. MENDELSOHN, D. H. ELLIS, D.G. SMITH, C. MEYBURG, AND A. C. KEMP. 1995.Year-round movements of a Wahlberg’s Eagle Aq-uila wahlbergi tracked by satellite. Ostrich 66:135–140.

MEYBURG, B.-U., W. SCHELLER, AND C. MEYBURG.2000. Migration and wintering of the Lesser Spot-ted Eagle Aquila pomarina: a study by means ofsatellite telemetry. Global Environmental Re-search 4:183–198.

MEYBURG, B.-U., J. MATTHES, AND C. MEYBURG. 2002.Satellite-tracked Lesser Spotted Eagle avoidscrossing water at the Gulf of Suez. British Birds95:372–376.

MIKHELSON, K. A. 1982. Steppe Eagle, Aquila nipa-lensis, p. 92–95. In K. A. Mikhelson and Y. A.Viksne [EDS.], Migrations of birds of eastern Eu-rope and northern Asia. Falconiformes-Gruifor-mes. ‘‘Nauka’’ Press, Moscow.

OSTROWSKI, S., E. FROMONT, AND B.-U. MEYBURG.2001. A capture technique for wintering and mi-grating Steppe Eagles in south-western Saudi Ara-bia. Wildlife Society Bulletin 29:265–268.

SHIRIHAI, H., R. YOSEF, D. ALON, G. M. KIRWAN, AND

R. SPAAR. 2000. Raptor migration in Israel and theMiddle East: a summary of 30 years of field re-search. International Birding and Research Center,Eilat, Israel.

SHOBRAK, M. 1996. Ecology of the Lappet-faced Vul-ture Torgos tracheliotus in Saudi Arabia. Ph.D.dissertation, University of Glasgow, Glasgow,UK.

SPAAR, R., AND B. BRUDERER. 1996. Soaring migrationof Steppe Eagles Aquila nipalensis in southern Is-rael: flight behaviour under various wind and ther-mal conditions. Journal of Avian Biology 27:289–301.

STEYN, P. 1982. Birds of prey in southern Africa. DavidPhilip Publishers, Cape Town.

UETA, M., F. SATO, E. G. LOBKOV, AND N. MITZA. 1998.Migration route of White-tailed Sea Eagles Ha-liaeetus albicilla in northeastern Asia. Ibis 140:684–686.

UETA, M., F. SATO, H. NAKAGAWA, AND N. MITA. 2000.Migration routes and differences of migrationschedule between adult and young Steller’s SeaEagles Haliaeetus pelagicus. Ibis 142:35–39.

WEATHERBASE [ONLINE]. 2002. Historical weather forDjibouti. ^http://www.weatherbase.com& (20 Jan-uary 2003).

MIGRATION ROUTES OF STEPPE EAGLES 227

WELCH, G. R., AND H. J. WELCH. 1988. The autumnmigration of raptors and other soaring birds acrossthe Bab-el-Mandeb straits. Sandgrouse 10:26–50.

WELCH, G., AND H. WELCH 1989. Autumn migrationacross the Bab-el-Mandeb Straits, p. 123–126. InB.-U. Meyburg and R. D. Chancellor [EDS.], Rap-tors in the modern world. World Working Groupon Birds of Prey, Berlin.

WELCH, G., AND H. WELCH. 1991a. The autumn mi-gration of the Steppe Eagle Aquila nipalensis.Sandgrouse 13:24–33.

WELCH, G., AND H. WELCH. 1991b. Spring raptor ob-servations from Djibouti. Ornithological Societyof the Middle East Bulletin 26:25–27.

YOM-TOV, Y. 1984. On the difference between thespring and autumn migrations in Eilat, southernIsrael. Ringing and Migration 5:141–144.

YOM-TOV, Y. 1988. Bird migration in Israel, p. 497–514. In Y. Yom-Tov and E. Tchernov [EDS.], Thezoogeography of Israel. Dr. W. Junk Publishers,Dordrecht, The Netherlands.

ZALLES, J. I., AND K. L. BILDSTEIN [EDS.]. 2000. Raptorwatch: a global directory of raptor migration sites.BirdLife International and Hawk Mountain Sanc-tuary, Cambridge, UK.

ZU-ARETZ, S., AND Y. LESHEM. 1983. The sea—a trapfor gliding birds. Torgos 5:16–17.