Breeding and mortality of OrientalWhite-backed Vulture Gyps bengalensis tn

Bird Conservation International (zrnz) rz:3;r-12.6 @ Birdlife International 2oo2IX)I:"10.10'17 /il959270902002198 Printed in the United Kinqdom

Punjab Province, PakistanMARTIN G ILBERT, MUNIR Z . V IRANI , R ICHARD T . WATSON, I .L INDSAY OAKS, PATRICK C . BENSON, ALEEM A . KHAN, SHAKEELAHMED, JAMSHED CHAUDHRY, MUHAMMAD ARSHAD, SHAHIDMAHMOOD and QASWAR ALI SHAH

Summary

Populations of Oriental White-backed Vulture Gyps bengalensis and Long-billed VultureG. indicus declined in India between the mid r98os and late r99os. Regional reports fromIndia described declines of g5-tooVo across a wide area. This study was conducted toinvestigate the breeding success and pattern of mortality in two vulfure colonies(Dholewala and Changa Manga) within Punjab Province, Pakistan between Decemberzooo and June zoor. Breeding success was found to be 6z%o in Dholewala and 59Vo inChanga Manga. A total of 668 sick and dead vultures were collected of which 591 wereless than one month post mortem. No significant variation was found in the weeklymortality rate of adult and subadult vultures during the study period spanning winterthrough summer. A peak in mortality rate was observed during late April and early Maythat corresponded to mortality of newly fledged juveniles. Minimum annual mortalityrate in the adult breeding population was calculated to be rt.4Vo and r8.6Vo in Dholewalaand Changa Manga respectively. In a subsample of dead vultures (n = r8:) visceral goutwas found in SoVo of adults, 63Vo of subadults, rg%o of juveniles and, tSVo of nestlings.These mortality rates were consistent with a rapid population decline. Results imply thatthe mortality factor responsible for the decline in Gyps vultures described in India is alsopresent in Pakistan and will potentially lead to a population decline of a comparablemagnitude.

Introduction

Oriental White-backed Vulture Gyps bengalenesis has been described as the com-monest vulture in the Indian subcontinent, with a range extending to south Viet-nam and the Malay peninsula (Ali and Ripley 1968). The species is resident inPakistan and is "widely distributed throughout the provinces of Punjab, Sindand the broader valleys of North West Frontier Province" (Roberts r99r). Thespecies "prefers cultivated tracts with scattered trees and a high human popula-tion, being attracted to larger towns and cities where slaughterhouses and refuse

tips offer more opportunity for obtaining food" (Roberts r99r).Declines in Oriental White-backed Vulture populations were first reported in

India by Prakash (tggg) in Keoladeo National Park, Rajasthan. Maximum vulture

counts within the park declined by g6vo from 1985/;.986 (max. u8m�) to rggS/

t-

M. Gilbert et aI. 372

1999 (max. 86). Numbers of active nests showed a comparable de.cline of. 95Vo,with 353 nests located in r9B7-88 and just zo in ry98/t999. Breeding success wasshown to have fallen from SzVo of eggs laid in ry85/ 1986 to o7o in ry97 / 1998 andt998 / tggg (Prakash & Rahmanirggg). Numbers of Long-billed Vulture G. indicussighted in Keoladeo National Park also declined, from a maximum count of 816birds in ry85 / t986 to only z5 in ry98 / r.999 Grakash ry9$.

Reports from numerous sources presented in Birdlife International (zoor) sug-gested that the decline in populations of G. bengalensis and G. indicus were morewidespread, extending across much of peninsular India. Observations suggestthat the recently described Slender-billed Vulture G. tenuirostrls may also havebeen affected (Birdlife International zoor). A reduction in the population of othersympatric species of Old World vulture, namely Egyptian Vulture Neophronpercnopterus and Red-headed Vulture Sarcogyps caluus, have not been described,suggesting that the declines may be restricted to the genus Gyps.

A wide range of hypotheses have been proposed to explain a decline of thismagnitude and rate. These include a reduction in food availability, loss of suit-able nesting habitat, pesticide intoxication, deliberate poisoning, emerging infec-tious disease and calcium deficiency (Prakash 1999). Investigations into the causeof the declines have been underway since 1999 (Oaks et al. zm1 Cunningham efal. zoot, Pain zoor) and have largely suggested the involvement of an infectiousdisease or pesticide intoxication. Visceral gout was reported to be a "common

finding" of vulture post-mortem examination in India (Pain zoor). Avian visceralgout should not be regarded as a disease entity, but as a clinical sign of anysevere renal dysfunction (Lumeij ryg4) and may be considered an end stage in abroad range of infectious and non-infectious disease processes. As yet no singleunderlying factor has been identified to account for the visceral gout that hasbeen found in the vulture population.

This paper reports the first attempt to measure rates and identify patternsof mortality, and measure rates of reproduction in extant colonies of OrientalWhite-backed Vulture in Asia. Data are presented on breeding success and mor-tality patterns in Punjab Province,.Pakistan between mid December 2ooo and theend of fune zoor. Data are also presented on the occurrence of visceral gout inrelation to body condition and age class. A method of estimating mortality ratewithin breeding populations is described to draw spatial and temporal compar-isons in this and future studies.

Methods

Our field studies were conducted in Punjab Province of Pakistan at two mainsites (Dholewala and Changa Manga) and 13 peripheral sites (Figure r). Dhole-wala (3o" 3z' 48" N, Zoo 5z' 43" E) in Muzaffargar and Layyah Districts, is locatedeast of the Indus River and comprises elevated canal banks lined with matureSheesham Dalbergia sfsoo trees that provide a linear distribution of suitable vul-ture nest sites. The land use is mainly agricultural, with cotton and wheat theprincipal crops. Changa Manga (3f o4' 10" N, 73" 58' J9" E) is located in KasurDistrict (south of the city of Lahore) and is a forest plantation providing a two-dimensional distribution of suitable vulture nest sites. It covers an area of 5,mzha and contains a wide diversity of tree species including Sheesham and Kapok

Oriental White-bqcked Vulture in Pakistan

Figure r. Location of the two main and 13 peripheral study sites.

Bombax ceiba. Abundant herds of domestic livestock occur at both sites. Dhole-wala lies 75o km and Changa Manga 55o km north-west of Keoladeo NationalPark, India. Dholewala and Changa Manga are 35o km apart.

surveys began in mid December zooo (Dholewala) and January zoor (Changa

4

t

INNIA

I

SINOH

t '

, s l-l

{ of o lmkm

€f,

: - * - * - * ' ' r y

t | . r @ tt t

; o Minot study siteg River t- l

i I Maior study site Natbnal boundary i

t * Provincial bo{rndery ii * Torvn i

,v

trNDrA I

I

M. Gilbert et aI. 344

Ma-nga), when pairs were in late incubation and early nestling periods rqspect-ively' All nests in Dholewala colony were located and marked regardless of occu-Pancy. A subsample of nests was randomly selected from all nests located formonitoring through the season. Four study plots were selected in Changa Manga,and all vulture nests were located and marked for monitoring. Nest observationswere made from the ground, and an assessment of nesting u"tirrity was recorded.Nests were categorized as containing an adult vulture inlncubating or broodingposture, an adult vulture or vultures present on the nest platform but neithei-"":" t9 be brooding nor incubating, a confirmed sighting of a nestling, or asbeing devoid of signs of breeding activity.

Activities at the Dholewala study nest sites were recorded on r9 days betweenr3 December zooo and 4-May zoor and on z3-58 days (depending on the studyplot) in Changa Manga, between z january and, z3

-Muy ,oot. FJrty-four nesis

were randomly selected from the total number of nests in Dholewala to estimatea mean nestling period. These nests were climbed on two to six occasions, start-lng as soon as possible after hatching in January and February to estimate nest-ling ages.

Using Postupalsky's (:.g74) criteria for assessing the reproductive success ofraptors, an active nest was defined as a nest in which eggs hld been laid, whereasan occupied nest was one in which an egg need not have Leen laid, but a minimumof nest building must have taken place. Postupalsky stressed that nest snccess canonly be determined from the number of occupied nests that successfully fledgeyoung. He also stated that it is necessary to survey early in the breedinf *u"Jn:: lhrj all territorial pairs will be included in an a-ssessment of occupariy. Giventhat observations began during incubation it was not possible to determine occu-pancy and therefore nest success as defined in Postupalsky. Indeed, consideringthe high nest density and vulture iraffic at both sites, it is unlikelv that an accur-ate figure of occupancy could be obtained even if observations weie to corunenceprior to laying. For this reason we used the active nest as the primary unit withinthis study. In practical terms, postupalsky stated that nests *i.ur" eggs were laid,young were raised, or an adult was observed adopting a posture consistent withincubation are sufficiently diagnostic to identify an-active nest. Benson et al.(r99o) present a method which may be used to obtain an estimated maximumnumber of active nests within a colony, using the daily rate of active nest failurewithin the observation period. This method is the same as that offered by May-held G97) to analyse incomplete breeding data sets in passerine birds, anddefines breeding success-across a population in terms of the propo*ion of activenests that fledge successfully.

The date of nest failure oifl"dgi.,g was taken as the midpoint between the lastconfirmed observation of-activity, and the first date from which no activity wasobserved. Exceptions to this occurred when the date of fledging or failure couldbe determined more precisely (such as in an observea nea"girL). Fledging wastaken as the absence

-or u yontrg vulture from the nest platfir# on o, after the

first date on which a known fledgling was observed in ihe colony. Beyond thisdate it was possible to_ distinguisli between failed and fledged neits as nestlingswere well grown by this stage and dead nestlings could 6e observed from theground without difficultv.

we measured mortaliiy by methodicaily collecting and removing dead vul-

Oriental White-backed Vulture in Pakistan 345

tures from beneath known nests throughout the breeding season. Study siteswere searched for dead vultures on 1o5 (Dholewala) and rzr (Changa Manga)occasions between r ]anuary and 3o June zoor. Detailed necropsies were per-formed on those dead vultures where autolysis was not yet advanced. \Alherepossible, the cause of death was determined from gross examination, andsamples were collected for histological, microbiological and toxicological ana-lysis. Observation of white uric acid precipitate over the surface or within theparenchyma of the liver, the kidneys, the lungs, within the pericardium, joints(femoro-tibial, tarso-metatarsal, shoulder), or any one of these was consideredsufficient to indicate the presence of visceral gout. Houston Qg76) used a systemfor scoring body condition of dead vultures by assessing the extent of fat reservesin the omentum, mesentery and subcutaneous tissue. Houston's scoring tech-nique was simplified in this study and a bird was considered to possess omentalfat reserves if deposits of fat formed a definite mass that obscured the intestineswhen abdominal musculature was resected. Birds with less extensive fat depositswere recorded as negative for omental fat.

The rate of decomposition varied greatly throughout the observation period,being significantly higher during the hot months of April, May and June' Evid-ence of uric acid accumulation was seen to persist several days after decomposi-tion prevented detailed post-mortem examination. In cases where it was not pos-sible to conduct a more detailed post-mortem, the skin, musculature and bodywall were incised to expose the thoracic and abdominal viscera. An assessmentof visceral gout and omental fat was made in these birds as described above, andwhere possible the sex of birds was determined. No assessment of articular goutwas made in these birds.

We recorded the age class of dead vultures using plumage characteristics foradult (full adult plumage), subadult (one year to adult), juvenile (fledging to oneyear), and nestling. We estimated time since death in periods of o-r day, z-7days, &-3o days, and > 3r days. The visual assessment of decomposition, and thefrequency of visits to the site, ensured that times since death were categorizedinto these broad classes with reasonable accuracy. Vulture carcasses wereremoved from the site, or where this was not possible, carcasses were buried, toavoid double counting. An estimation of annual mortality rate was calculated bydividing the number of dead vultures (DV) located within the study plots duringthe observation period (OP in days) by the number of breeding individuals(based on the number of active nests (AM at the start of the observation period)and extrapolating this over a 365 day period; using the formula:

D V / ( A N x z ) . . _ , _oP

-x3b5

In our analyses of mortality rate we excluded all dead vultures classed as 2-7days post mortem located less than one week after the start of the observationperiod, all 8-3o day vultures located less than one month after the start of theobservation period, all birds > 3r days post mortem and birds for which posfmortem interval was unknown. Mortality rates were calculated for adults alone,and adults and subadults combined.

Patagial wing tags were fitted to 47 nestlings from zz March to 4 April inDholewala following the method outlined in Wallace et aI. (r9{h). A further z4

M. Gilbert et aI.

nestlings werebetween r and

3a6

tagged at a second colony, Toawala (3oo4o'25'N, 7o"55'tt" E)z April. At both sites the identity and lncation of wing-tagged

nestlings were recorded whenever they were encountered or carcasses recovered(Dholewala 53 and Toawala 68 observation days). No attempt was made to re-locate wing-tagged fledglings beyond the immediate vicinity of the nesting col-onies.

Results

Breeding success

A total of 3,98o nests were located across the two primary and 13 peripheralstudy sites. of these 2,43c^ were considered active, and at i,293 no a"lirrity *utobserved. In Changa Manga 1,051 nests were located, of which 75g ( TzToi wereactive. Nesting density in Changa Manga was o.15 active nests per hectare (n =758 nests within 5,ooz ha of forest). In Dholewara 6o7 nests *ere located., ofwhich breeding activity was recorded in 4r3 6gV"). Nesting density in Dholewalawas 19.1 active nests per kilometre of linear transect (n = 4r3 along zr.6 km). InDholewala ggvo of nests were built on sheesham. In comparisor,l th"r" was agreater diversity of nest trees in Changa Manga where biids nested in 14 treespecies (Sheesham 4Vo, Albizzia lebbeck t7Vo, Calistemon lanceolatum r47o, Ficus sp.4vo, Terminalia sp. 3vo, Trewin sp. 3To, Bischafia sp. tvo, Morus nigra t%, Angophorasp._t72 -Celrein sp.t%o, Greuillea sp.rVo, pinus sp.t%o, Salmalin sp-.tTo).- In Dholewala 3o7 nests were selected foi monitoring oi breeding successbetween December zooo and May zoor. of these, 47 *"i" considered active inDecember, 66 showed no activity and status could not be determined for four.Durinq the 44 day observation period fledging was confirmed in r53, and nest-ing failure in 84 nests. Daily rate of nest fiituie during the observation periodwas o.583 nests per day. The first fledgling was observed on z3 March, uttd *""r,fledging date was 19 April. of 44 nests viiitea to assess nestling period, 36 wereactive in December. Nestlings fledged from z5 of these. Mean nestling periodwas determined to be ro4 days. The incubation period was assumed io be 56days as stated for the closely related African White-backed Vulture G. africaius(Mundy et al. ry92). The estimated number of active nests at laying was 246, withsuccess of active nests calculated to be 6z%. This was calculaied using the dailyfailure rate during the observation period.

we monitored r89 nests for breeding success in changa Manga between janu-ary-and May zoor. of these, rr7 were active in January, rr showed no activityand status could not be determined for one. Nestlingi fleaged from 116 sites,and 6r failures were confirmed. Mean fledging date wis z4 April. The estimatednumber of active nests at laying was ag\, with a breeding r.t.i"rt of 5gvo.

Mortality

We collected 668 dead, sick and injured Oriental White-backed Vultures from allour study sites during the study period. one dead adult Eurasian Griffon G.fulous was collected on r7 June, buf was too decomposed to indicate the cause ofdeath. Dholewala and Changa Manga accounted for ry6 (263%) and rg6 (27.g%)

Orientsl White-backed Vulture in Pakistqn 347

Table r. Number and age distribution of dead, sick and injured vultures collected within study

transects (excluding all birds that died prior to the study period) in Dholewala and Changa Manga,

Pakistan (January to July zoor)

Adults(%)

Subadults(Vo)

Fledglings(7")

Nestlings(Vo)

Total(Vo)

DholewalaChanga Manga

z6 (26.8)z5 Q69)

9 \W)z3 \24.7)

45 Q6.4)39 Qt.g)

17 $7.5)6 (6.5)

97 Goo)93 (roo)

Table z. Estimated time since death of dead vultures collected within study transects (excluding allbirds that died prior to the study period) in Dholewala and Changa Manga, Pakistan $anuary to

July zoor)

o-r day z-7 days 8-3o days >3rdays Sick/injured Unknown

DholewalaChanga Manga 10

2

866475

1

2

1

o

dead vultures located respectively. The proportion of each age class known tohave died within the study plots during the study period is summarized in Tabler. Post-mortem intervals of vultures located are summarised in Table z. Adultannual mortality rate was calculated to be rr.4Vo in Dholewala, and 18.67o inChanga Manga. Combined adult and subadult mortality rate were calculated tobe t537o in Dholewala and 36:zVo in Changa Manga.

The weekly adult/subadult mortality rate using birds less than seven dayspostmortem was 1.54 (SE = o.33) and t77 (SE = o.z6) vultures per week in'Dholewalaand Changa Manga respectively (Table 3, Figure z). Dead fledglings began toappear from weeks 14 pnd r7 in pholewala and Changa Manga respectivelyGigure 3). Weekly fledgling mort4lity rate during the final 13 weeks of the studyperiod was calculated ttr be 4.69 (SE = z.3o) in Dholewala and 3.69 (SE = r.43) in-hanga Mgnga. Fledgling mortality in the immediate vicinity of study nests was27.5% in Dholewala (n = t53 successful nests) during 13 weeks and 33.6Vo inChanga Manga (n = 116 successful nests) during ro weeks.

Eighteen $$Vi tagged fledglings were resighted in Dholewala in 53 days ofobservation and 14 (3o7o) were recovered dead; trvo of these prior to fledging(47o) andseveir (r57o) from their natal nes-ts during the post-fledging dependenceperied (PFDP). One bird resighted in pholewala was later found dead on itsnatal nest and so is included in the mortality figrle stated. Ten (37Vo) tagged

Table 3. Weekly and annual mortality rate of Oriental White-backed Vultures at Dholewala and

Charrg"a Mangai Pakistan

Study Weekly Weekly Annual Annualperiod adult/ fledgling mortality rate mortality rate(weeks) subadult mortality rate (adults only) (adult,/

mortality rate (SEF subadult(SE) combined)

DholewalaChanga Manga

z6 r.54 @33)z6 r.77 @.26)

4.69 Q3t) l.4Vo

3.69 3.q) r8.6Vot53Vo

36.z%o

'Over 13 weeks post fledging.

378

rowO C M

I 3 5 7 9 l l 13 15 t7 19 2t 23 25

Weeks commencing January 2001

Figure z. Dead adult and subadurt oriental white-backed vultures found per week atDholewala (DW) (z = 4r) and Changa Manga (CM) (n = +ol, p"ririur,, arring the studyperiod.

aot)g ? {

o0()= z odc)

lHo 1 5{)

e l 0

rDWtrcM

r4 t5 t6 t7 18 19 20 21 22 23 24 25 26

Weeks commencing January 2001

1ry:".3. D-uld fledgling oriental white-backed vultures found per week at Dhorewala(DW) (n = 6r) and Changa Manga (n = 4g), pakistan, during the study period.

Orientai l lhite-backed Vulture in Pakistan 319

fledglings were resigliteri in Toawala during 68 observation days. Six (zz%o) weterecovered dead, all of which had fledged, and were located away from their natalnest. Across all sites 9.gVo of tagged fledglings (n = zo) and 8.z%o of non-taggedfledglings (n = zt)) were recovered from their natal nests.

Gout

An assessment of gout was made in r85 dead vultures (45 adults, z4 subadults,

93 fledglings, and z3 nestlings). Of these, 7z QgVo) had signs of visceral gout.Gout was found inSo% of adults, 637o of subadults, rg%o of fledglings, and 9%of nestlings. Cases of gout were observed in every month between Decemberand June. An assessment of sex was made in 49 dead vulfures, of which z6 weremale and z3 female. Twenty males and 15 females showed signs of visceral gout.Omental fat reserves were found in 8r% of birds with visceral gout (n = 7z) and8t% of birds with no signs of gout (n = rrJ). Detailed necropsies were conductedon 36 vultures. The primary cause of death was established in tz cases, only oneof which also had gout (Table 4). A further case was found to have a liver abscesssuggestive of avian tuberculosis, but carcass decomposition precluded collectionof samples. Primary cause of death could not be established in zz cases, zo ofwhich had visceral gout (Table 4).

Discussion

Old World vultures conform to a life history strategy typical of large raptors andseabirds (Wynne-Edwards tg55, Amadon 1964, Newton agTg,Piper et al. ry8t).They are long-lived, reproduce slowly, and adult survival is high in comparisonwith smaller birds. Among most animals including raptors, there is a strongpositive correlation between body weight and adult survival, such that heavierspecies live longer, and are subiect to lower mortality rates (Newton ry79). Oti'ental White-backed Vultures are comparable in weight to Bald Eagles HaliaeetusIeucocephnlus, and would be expected to have a similar annual adult survival rateof >o.9r (Newton tg7g, Bowrnan et al. 1995). The stability of a non-migratoryraptor population is generally dependent on three main factors: adult survival,non-adult survival and productivity (Watson 1990). For large, long-lived raptors,population stability is most sensitive to changes in adult survival, such that asmall increase in adult mortality (r - survival) relative to a similar change innon-adult survival witl rapidly exceed the maximum productivity and force apopulation into decline.

Quantifying mortality rates in raptors is notoriously difficult. Recognizedmethods are broadly based on the rate of decline at which dead birds are reco-vered from a marked cohort (Newton 1979). Such studies are conducted over aperiod extending to years, and accuracy will be directly influenced by the pro-portion of the population marked, and the recovery rate of dead birds. In thecontext of vultures in south Asia, such long-term studies are impractical con-sidering the rapid rate of decline observed and the urgency for remedial action.An alternative method uses plumage characters to assess age class distributionwithin a population and assumes any reduction in representation of each ageclass is due to mortality, and the population is stable. This method is only useful

M. Gilbert et al.)20

Table 4 Oriental white-backed Vultures case summaries of post-mortem examinations conductedbetween November zooo and June zoor

Case Dateno.

Age Sexclass

Site Body Gout Provisional diagnosisfat

1

2

3+56

78

910

1 L

L2

73L475t6L 7r81"9

21

) )234

25z627zB29

30373233343536

zr Nov oozz Nov ooz3 Nov ooz Dec oo16 Dec oo16 Dec oo16 Dec oo17 Dec oo17 Dec rnzr Dec oozz Dec oo3 Jan or

3 Jan or

9 Jan or14 Jan orr5 fan or3r Jan orr Feb orr Feb or5 Feb or5 Feb orrz Feb orrz Feb orr8 Feb or18 Feb orr8 Feb orzo Feb or5 Mar orz7 Mat or1 Apr 01r Apr or5 May orro |un orro Jun orr3 |un or13 Jun or

Changa MangaLahore

ChichawatniChanga MangaKatoraKatoraKatoraKatoraChanga MangaDholewalaDholewalaDholewala

DholewalaChanga MangaChanga MangaChanga MangaToawalaToawalaDholewalaChanga MangaChanga MangaToawala

Changa MangaDholewalaDholewalaDholewalaDholewalaChanga MangaToawalaToawalaToawalaDholewalaChanga MangaChanga MangaChanga MangaChanga Manga

AdultAdultJuvenileAdultAdultAdultAdultAdultAdultAdultAdultSub Adultr weekSub AdultSub AdultAdultSub Adult3-4 weeksz weeksAdultAdultAdultAdultAdultAdultAdultAdultSub Adult7o days8o days9o daysFledglingFledglingSubadultAdultAdult

Yes YesYes NoYes NoYes YesYes YesYes YesYes YesYes Yes? N OYes YesYes YesNo NoNo NoNo NoYes YesYes YesYes NoNo NoNo NoYes YesYes YesNo EarlyYes YesYes YesYes YesYes YesYes NoYes NoYes NoYes NoYes NoPoor NoYes Yes? YesYes YesYes Yes

MFFMMM

M?MMM?FMFF??FFFF,;?

i???MMFFM

Op"tWire collision, traumaHit by car, traumaAvian tuberculosisOpenOpenOpe.OpenHuman persecutionOp.nOp".Open (CNS signs)Fell from nest, traumaLead poisoningOp"^Op".GunshotFell from nest, traumaFell from nest, traumaOpenOpenDiarrhoea, openOpenOpenOpenOpu.Liver abscess, TB probNormal ? (captured)Fell from nest, traumaFell from nest, traumaFell from nest. traumaDiarrhoea, openOpenOp"tOpenOpe.

where age class can be assessed in the field with confidence. Although it is pos-sible to subdivide a population of Oriental White-backed Vultures int"o broad ageclasses (juvenile, subadult, adult), knowledge of the moult sequence of the sp"e_cies is insufficient to accurately relate these plumages to age. witn tn" rrrrrrrrruilylarge number of dead vultures located in thls studv it is-reasonable to assumlthat the population is not stable. For these r"uro.,", and because mortality wasso high that it was practical to collect dead birds in statistically useful rrrr*b"rr,we estimated annual mortality rate within the breeding populition by assumingall birds collected from beneath nests were part of the bieeding population, replresented by the number of active nests multiplied by two. Calcriliting the mortil-ity rate was therefore limited to only the breeding population. our intention was

Oriental White-backed Vulture in Pakistan )24

to quantify vulture mortality as best we could, understanding the limitationsimposed on the value by our assumptions, and with sufficient precision to pro-vide an index of mortality rate that might be used in comparisons between geo-graphical areas and with subsequent breeding seasons.

All methods of quantifying the mortality rate of wild populations of raptorsare subject to bias and assumptions that may affect the validity of the result(Newton 7979).We made three important assumptions. First, we assumed thatall dead birds from our subsample of the breeding population were located. Inpractice this was unlikely to be the case, as only birds dying in the immediatevicit ity of their nests were found. Removal and consumption of dead birds byother icavengers such as dogs, almost certainly reduced the number located stillfurther. Under-recording of dead birds probably led to an underestimate of thetrue annual mortality rate of the breeding population. Second, by extrapolatingmortality measured during the study period to annual mortality, we assumedthat mortality rate remained constant. We have shown that during the six monthsbetween fanuary and July there was no significant variation in the weekly mor-tality rate of adult and subadult vultures, so this assumption may be colrect.However, the identification of a seasonal pattern in the mortality of Asian vul-tures would have a significant impact on our understanding of its epidemiolog-ical characteristics, so the subject warrants further study throughout the year.This is difficult to achieve given that our method depends on the breeding PoPu-lation being tied to the breeding site, which occurs for only eight months of theyear. Third, we assumed that all dead vultures found in the vicinity of nestswere part of the breeding population. Oriental White-backed Vultures normallyexhibit deferred maturity, so annual mortality rates calculated using the numberof dead adults would normally be most accurate. However, an unknown ProPor-tion of the breeding population in Pakistan had not attained adult plumage andmay be considered "immature breeders". Hence, mortality rates calculated usingadult vultures only will underestimate mortality rate in the breeding population.Annual mortality rates calculated using a combination of adult and subadultvultures will account for "immature breeders", but may also include non-breeding subadults, and should probably be considered an upper range forannual mortality rate within the breeding population.

The range of annual mortality,rate was t.4Vo (adults only) to t537o (adults

and subadults combined) in Dholewala and t8.6% (adults only) to 36.z%o (adults

and subadults combined) in Changa Manga. Given our assumPtions, we suggestthe adult-only mortality rate considerably underestimates the true annual mortal-ity of the breeding population. The range of annual mortality rates we presentaie higher than known and calculated mortality rates in stable populations ofother raptor species of comparable body weight, demonstrating that the coloniesstudied were in a state of rapid decline during the study period. Studies at thesesites are ongoing, and will provide data that will test the accuracy of these predic-tions.

sharma GgTo) recorded the breeding success of oriental white-backed vul-tures near |odhpur, Rajasthan, as g6Vo, though the definition of breeding successused was not stated. In Keoladeo, Prakash (rggg) indicated breeding success ratesof SzVo in ry85/:'986 (n = 244) and o in the aggT/rgg9 (n = z5) and ry98/99 (n =

20) seasons. He defined breeding success as the number of birds raised to

M. Gilbert et al. 322

fledging per egg laid, so his results can be directly compared to this study. Heindicated zo%o of nest failure occurred prior to laying, zoVo during incubationand 6o% during the nestling period in ry97/r998 and 6o% prior to Laying, ro%oduring incubation and 3o% during the nestling period in ry98/ t999. Though ourvalues of 6z.tVo and. 58.8Vo in Dholewala and Changa Manga respectively inzooo/ zutt are not as high as the pre-decline figures of Sharma ggTo) or Prakash(1999) they are still far from the total disappearance of breeding activitiesobserved in Keoladeo in the late r99os.

It is important to recognize that our breeding success figures have been extrapol-ated using the known failure rate during the observation period to predict thenumber of active nests at laying. This method assumes that the rate of nest failureremained constant from laying to fledging. This introduces a potential source oferror in our study as observations were largely made during the nestling period.The post-decline figures presented by Prakash (tggg) indicated three times as manynests failed during the nestling period as during incubation in both tggT / 1998, andin ry98/ t999. Assuming that the incubation period in Oriental White-backed Vul-ture is equal to the 56 day incubation period seen in African White-backed Vulture(Mundy et aI. tggz), and that our estimate of mean nestling period of ro4 days isaccurate, then Prakash's nest failure rate during the incubation period was r.5times greater than during the nestling period. A difference in failure rate of thismagnitude in our study colonies would translate to an amended breeding successof 54.zVo in Changa Manga and 57.57o in Dholewala. At the opposite extreme, werewe to assurne that no nests failed prior to the start of the study, failure rate withinour colonies would be 64.6Vo in Dholewala and 65.5Vo in Changa Manga. In bothcases it is clear that breeding success in the colonies studied is lower than the pre-decline data of Sharma (r97o) and Prakash Gggg).

Our findings of visceral gout in 78Vo oI adult (n = 45) and 63Vo of subadult (n =24) vultures examined is consistent with preliminary findings of post-mortemexaminations conducted in India (Pain zoor). It is possible that vultures areunusually susceptible to developing this condition, and that all cases need notbe related. However, of zr birds with visceral gout where a detailed post-mortemexamination could be made, a primary diagnosis was only possible in one case(Case 4: avian tuberculosis). It is beyond the scope of this paper to cover thedetailed clinical findings of the birds examined and this will be described in fullin a separate publication. Suffice to say that lesions associated with the precipita-tion of uric acid were the only consistent findings made during the detailednecropsy of zo birds with visceral gout. The lack of a firm diagnosis or lesionsunrelated to uric acid in the majority of birds with gout was consistent andcharacteristic of the majority of dead vultures, but so far has done nothing toreveal the cause of mortality. This observation supports the hypothesis that casesof visceral gout are related and hence would suggest that the mortality factorresponsible for the decline in India is also present in Pakistan.

The body condition scoring system used by Houston (1976) has been greatlysimplified for use in this study. Houston's system defines nine classes of bodycondition based on deposits of fat in the mesentery, omentum and subcutaneoustissue. In this study we have scored birds with moderate to abundant omentalfat reServes as "positive" for omental fat, which would conespond to a bodycondition score of 69 on Houston's scale. Therefore, it should not be assumed

Oriental White-backed Vulture in Pakistan 323

thai birds scored as "negative" in our study are completely emaciated, butsimply indicate that these birds have a body condition that would be consideredmoderate to poor.

Moderate to abundant reserves of omental fat were found in 8t% of deadbirds. This supports the assessment of Prakasn (rggg) that the observed mortalityin Gyps vultures is unrelated to the availability of food. Prakash also states thatprior to death vultures would remain immobile for "more than 3o days". It isclear from our study that this is not the case for the majority of vultures thatultimately die. Moderate to abundant levels of omental fat are unlikely to befound in a bird that remains immobile for a period in excess of one month, andwould suggest that death in the majority of cases in this study followed a moreacute clinical course.

Visceral gout was also recorded in juveniles and nestlings, although at a lowerrate than was observed in adults and subadults. This may be due to a lowersusceptibility to the condition than is seen in adult and subadult birds, a lowerexposure to factors leading to visceral gout, or a low prevalence of gout casesrelative to the incidence of other age-related mortality factors. It is likely thatdata on the occurrence of gout in nestlings has suffered from bias, with an over-representation of nestlings that have fallen from their nests. Mortality rate infledglings greatly exceeded that of adult and subadult birds, representing 43Voof birds that died during the study period. Mortality of fledglings was concen-trated in a r3 week period between April and June. Comparison with other spe-cies of Gyqs vultures suggests that high fledgling mortality may not be unusualfor the species (Houston r974,Piper et al. rg8l^, Robertson 1985, Benson eooo).

Benson (zooo) stated that fledging corresponds to one of the periods of highestmortality for all birds, although this is very difficult to quantify in vultures dueto the long distances birds may cover before dying. Studies of mortality in CapeVulture G, coprotheres based on recovery of birds banded as nestlings suggestedmortality rates in the first year in excess of 5o%o (Houston ry7Q to 83Vo (Piper etat. ry8i. studies based on resighting of marked Cape vultures at Potberg inSouth Africa (Robertson 1984) also predicted that first year mortality wasbetween 7t and 8r% (assuming that all birds that failed to return to the natalcolony were dead). Parental care in Old World vultures continues for anextended period after fledging. During the post-feldging dependence period(PFDP) juveniles must return to the nest in order to be fed by their parents(Robertson 1985). Mortality in the PFDP is largely a consequence of the inabilityof young vultures to return to the nest (Benson zooo)'

Our finding that g.9Vo of dead wing-tagged fledglings (n = zo) and 8z% ofnon-wing-tagged fledglings (r = 2rg) in this study were found in their natalnests represents further evidence that colonies surveyed were subject to unusualpatterns of mortality. This group of fledglings was found in a position that wouldnormally be considered compatible with survival. Data are insufficient to predictwhether fledgling mortality within the nest was related to mortality of the parentbirds or through gout-related death of the fledglings, although of four fledglingsrecovered from nests where an assessment of gout was made, the condition wasidentified in only one bird.

Although visceral gout was only found in a minority of juveniles and nestlings,the fact that it was found at all should be considered significant. If the hypothesis

M. Gilbert et aI. j24

that gout-related deaths are connected is correct, then the presence of the condi-tion in juveniles and nestlings would suggest that the mortality factor responsiblefor the condition is either vertically or horizontally transmissible from adults tooffspring, or is inherently present in the vicinity of the colony itself.

In conclusion, this study supports the hypothesis that the mortality factorresponsible for the decline in vulture populations in India is also present inPunjab Province, Pakistan. Breeding success at two colonies in Punjab is some-what lower than pre-decline figures reported from India. Despite an elevatedmortality rate and a depression in breeding success, large colonies are still pre-sent in Punjab, Pakistan. This may be due to a lower exposure rate to the mortal-ity factor than in India, or may reflect a westward expansion of a hypotheticalinfectious agent or other cause, which has yet to lead to declines of the magni-tude observed in India. Continued monitoring will be required to test the hypo'thesis that observed mortality rate will lead to a reduction in the breeding popu-lation in future seasons. The need to continue the diagnostic work alreadyunderway should be considered of utmost importance. Priority should be givento identifying the mortality factor and understanding the manner of its interac-tion with the vulture population, with the ultimate aim of reducing exposure orincreasing resistance of the remaining wild birds, sufficient to allow the popula-tion to recover.

Acknowledgements

Funding for this proiect was received from the Gordon and Betty Moore Founda-tion, UN Foundation, The Walt Disney Company Foundation ConservationAwards, and the Zoological Society of San Diego. We would like to thank thefollowing individuals and organization's for their partnership, help andcooperation: in Pakistan, the Ornithological Society of Pakistan (OSP), BrigadierMukhtar Ahmed, Punjab Department of Wildlife and Parks, Lahore Zoo,National Council for the Conservation of Wildlife (NCCW), B.Z. Multan Univer-sity, University of Agriculture at Faisalabad, Sind Wildlife Management Board,Zoological Survey Department Pakistan, VVWF (Pakistan) and Pakistan Museumof Natural History; in Nepal, Bird Conservation Nepal (BCN), HimalayanNature, Koshi Camp, and Department of National Parks and Wildlife Conserva-tion (DNPWC). In the USA, The Zoological Society of San Diego, WashingtonState University and the Bodega Bay Institute. We alio thank Asim for his invalu-able assistance and Ramzan for his tireless and dedicated field efforts.

References

Ali, S. and Ripley, S. D. (1968) Handbook of the birds of Indin and Pakistan together utith thoseof Nepal, Bhutan and Ceylon, L Diaers to hawks. Oxford: Oxford University Press.

Amadon, D. (rg6q) The evolution of low reproductive rates in birds. Eaolution r.8: ro5-

Benson, P. C. (zooo) Causes of Cape Vulture Gyps coprotheres mortality at the Kransbergcolony: a 1-T year update. In R. D. Chancellor and B.-U. Meyburg, eds. Raptors at Risk.Proceedings of the V World Conferance on Birds of Prey and Otols. Midrand, fohannesburg,South Africa, 4-rr August 1998.

Oriental White-backed Vulture in Pakistan 325

Benson, P. C., Tarboton, W. R., Allan, D. G. and Dobbs, J. C. (rggo) The breeding status

of the Cape Vulture in the Transvaal during r98o-r985 Ostrich 6t: r34-t4z'

Birdlife International (zooi Threatened birds of Asin: the Birdlife International Red Data Book.

Cambridge, U.K.: Birdlife International.Bowman, T. D, S.h"*pt P. F. and Bernatowicz, J. A. (rggs) Bald Eagle survival and

population dynamics in Alaska after the Exxon Valdez oil spill. I. Wildl. Manag. 59t 377-

324.Cunningham, A., Prakash, V., Ghalsasi, G. R. and Pain, D. (zoor) Investigating the cause

of catistrophic declines in Asian Griffon Vultures, Gyps indicus and G. bengalensis. lnT.

Katzner and f. larry-;ones, eds. Reports from the workshop on Indian Gyps Vultures.

4th Eurasian Congress on Raptors, Sevilla, Spain, September 2@r'Houston, D. C. (rgZD Mortality of the Cape Vulture. Ostrich 45: 574zHouston, D. C. (t976) Breeding of the White-backed and Ri.ippell's Griffon Vultures, Gyps

africanus and G. rueppellii. lbis rr8:. 14-39.Lumeij, !.T. (rggQ Nephrology. PP.SI8-SSS in B. W. Ritchie, G. f. Harrison L' R' and

Harrison, eds. Aaian meilicine: principtes and application Lake Worth: Wingers Publishing'

Mayfield, H. F. (rg7) suggestions for calculating nest success. wilson BuIl. 87: 456-465.Mundy, P., Butchart, D., Ledger, J. A. and Piper, s. fi. (t992) The zsultures of Africa. south

Africa: Acorn Books.Newton, l. (ag7il Population ecology of raptors' Berkhamsted, U.K.: Poyser.

Oaks, J. L., Rideout, B. A., Gilbert, M., Watson, R' T., Virani, M. Z. and Khan, A' A' (zoor)

Summary of diagnostic investigation into vulture mortality: Punjab Province, Pakistan,

2o(y)-2ocrL.In T. Katzner and ). Parry-Jones, eds. Reports from the workshop on Indian

Gyps Vultures. 4th Eurasian Congress on Raptors, sevilla, spain, septembet 2$r-t.

Pain, D. J. (zoor) Conservation of critically endangered Gyps vultures in India. In L.

Bennun and M. Virani, eds. Responding to the Asian vulture crisis: planning for vulture

monitoring and conservation in Kenya. Proceedings and recommendations of a seminar

and workshop held at the National Museums of Kenya, May zoor. NMK Ornithology

Research report 41, June zoor.piper, S. E., Mlndy, p. |. and Ledger, J. A. (rg8r) Estimates of survival in the Cape Vulture

Gyps corprotheres. l. of Anim. EcoI. 5o:.8t5-825.Postupalsky S. (rgZ+) Raptor reproductive success: some problems with methods, criteria,

and terminology. Pp. ir-y in F. N. Hamerstrom, Jr., B. E. Harrell and R. R. Olendorff,

eds, Managemint of raptors, Proceedings of the conference on raptol consmntion techniqun,

Fort Collins, CO: zz-24 March, tg73 (patt 4). Raptor Research rePort 2'

Prakash, V. (rq99) Status of vultures in Keoladeo National Park, Bharatpur, Rajasthan

with special reference to population crash in Gyps species. l. Bombay Nat. Hist' Soc. $:

365-78'ftuUtn, V. and Rahmani, A. R. (rqg) Notes about the decline of Indian vultures with

particular reference to Keoladeo National Park. Vulture News 4t:6-t3'pratcastr, V. (zoor) Status and distribution of vultures in India with special reference to

the population crash in Gyps species of vultures. In T. Katzner and |. Parry-jones, eds.

Repbrti from the workshop on Indian Gyps vultures. 4th Eurasian Congress on Raptors,

Sevilla, Spain, September 2ooo.Roberts, T. l. (rS9ti The birds of Pakistan, r. Non-passeriformes. Oxford: Oxford University

Press.Robertson,A.S.(1984)AspectsofthepopulationdynamicsofCapeVulturesintheCape

Province. Ostrich 55; t9Gzo6.Robertson, A. S. (rq8S) Observations on the post-fledging dependence period of Cape

vultures. Ostrich 56: 5&-66.Sharma, 1.,K. (tg7o) Breeding of Indian White-backed Vulture at Jodhpur. Ostrich 4ri 2o5-

207.

L

M. Gilbert et al.

wallace, M. P., Parker, P. G. and remple, s. A 1r93ri) An evaluation of patagial markersfor cathartid vultures. I. Field Orni.5r:3o9-428.

Watson, R. T. (r99o) Population dynamics of the Bateleur in the Kruger National Park.Ostrich 6t: 5-rz.

Wynne-Edwards, V. C. (1955) Low reproductive rates in birds, especially sea-birds. Pp.540-547 in Acta XI. International Ornithological Congress, 1954.

MARTIN GILBERT, MUNIR Z. VIRANI and RICHARD T. WATSON,The Peregrine Fund, 5658 West Flying Hawk Lane, Boise, Idaho, 8j7o9, llSA.

E -mail : mar t gilbert@y ahoo. com

I. LINDSAY OAKSDepartment of Veterinary Microbiology and Pathology, Washington State Llniaersity, Pullman,

W ashington 99 fi 4-7 o 4o US A.

PATRICK C. BENSONSchool of Animal, Plant and Enoironmental Sciences, Llniaersity of the Witwatersrand, Priaate

Bag j, P.O. WITS, zo5o, lohannesburg, Republic of South Africa.

ALEEM A. KHAN, SHAKEEL AHMED, IAMSHED CHAUDHRY, MUHAMMADARSHAD, SHAHID MAHMOOD and QASWAR ALI SHAH

Ornithological Society of Pakistan, rc9/D P.O Box 7j, Dera Ghazi Khan, Pakistan.

Received 15 fanuary zooz; revision accepted 5luly zooz

326