A brief history of the reintroduction of the Arabian oryx Oryx leucoryx into Oman 1980–1992

UNGULATES 81

Int. Zoo Yb. (1993) 3 2 81-90 0 The Zoological Society of London

A brief history of the reintroduction of the Arabian oryx

into Oman 1980-1 992 Oryx leucoryx

ANDREW SPALTON White Oryx Project, O@ce of the Adviser for Conservation of the Environment, Diwan of Royal Court, PO Box 246, Muscat, Oman

In 1972 the Arabian or White oryx Oryx leucoryx was extinct in the wild. Nine years earlier a captive ‘World Herd’ of nine oryx had been established in the USA (Fitter, 1982). Other captive herds existed in Arabia. In 1980 animals were brought back to the central desert of Oman for release on the Jiddat al-Harasis (Stanley Price, 1986).

MONITORING Monitoring of this population is primarily the responsibility of the Harasis ranger patrols. Between 1980 and 1986 most oryx

were seen daily and details of location, herd composition, births, deaths and sexual activity were recorded. Paternities were determined from observed mating. A small number of immigrant animals were marked but most were known by varia- tion in body or face markings, horn shape and general body shape (Stanley Price, 1989).

After 1986 sightings of oryx were less frequent and paternhies were increasingly determined by association. In 1990 a new monitoring strategy was adopted which enabled the collection of detailed informa-

Plate l. Arabian oryx Oryx leucoryx herd in the wild in the Jiddat al-Harasis.

82

tion for a sample of known animals and low-level monitoring of all others. To facilitate this some wild-born animals were ear tagged for individual recognition and by December 1992 32% of the popu- lation could be identified by ear notches, radio collars or ear tags.

Range use by individuals was calcu- lated using minimum convex polygons around known sightings (Nottrott, 1986). When calculating 8 territories the influence of large peripheral areas of low usage on an animal’s range was reduced by removing the outlying 5% of sightings.

EARLY YEARS 198e-1986 The period up to June 1986 is described in detail by Stanley Price (1989). The popu- lation grew at a mean annual rate (A) of 1.22 and by June 1986 numbered 31 in the wild (Fig. 1 ) .

Immigrants The first 5.5 immigrants were transferred from the USA to the release site at Yalooni, on the Jiddat al- Harasis, in 1980. A further 3.1 animals followed in 1981 and 0.3 in 1982. A single

arrived from the Shaumari Reserve in

no 1s

100

5c

live

@ d e s e r t born

surviving immigrants

I. I- . I 4 immigrants

Fig. 1. Numbers of Arabian oryx Oryx leucoryx alive at the end of each year and numbers of immigrants (solid bars) during the years 1!380-1992.

Jordan in 1984. These animals formed the core of two herds released to the wild in 1982 and 1984. A single S from the USA, unsuitable for release to the wild, was transferred to the Bait Barakah Breeding Centre (BBBC) near Muscat.

Fecundity and mortality There were 32 births of which three were stillborn (Stanley Price, 1988). The mean age at first calf for 99 born in the desert during this period was 36.36f8.77 (S.D.) months. Three immigrant animals, which had given birth in the USA or travelled pregnant to Oman, first calved at 25+ 1.54 months of age (Stanley Price, 1989) suggesting that desert conditions, even with animals receiving feed supple- ments during a long dry period, did not favour early conception. Immigrant 99 calving for the first time in Oman showed a breeding delay. Male breeding was dominated by three immigrant animals and although the oldest desert-born 3 was aged three-and-a-half years by June 1986, none of the young 83 had obtained a breeding position.

Of the 18 immigrants 39% had died by June 1986. Causes of death were varied though most were associated either with management activities and captive condi- tions in Oman or were the consequence of captive-bred animals proving to be not suited to life in the wild. For example, causes included physical injury following collision with fencing prior to release and bacterial infection in individuals which were scavengers of rubbish. Causes of the death of 24% of the desert-born animals were different suggesting more natural mortality factors. These included calf separation, infanticide and predation. Including still births, calf mortality to 30 days was 25% (Stanley Price, 1989).

Social and spatial organization The two released herds proved cohesive and stable with few changes during the early years. Each comprised adult and sub-adult 33 and 99 with following offspring and was of almost equal sex ratio. Dominance was

UNGULATES 83

hierarchical with a single adult bull controlling breeding.

Although a long dry period between 1983 and 1986 meant that the oryx spent extended periods close to the release site, where they received supplementary feed and had access to water, there was a slow expansion of range and by 1986 the first herd released had a total range of 2000 km2 (Stanley Price, 1988). Total range comprised a series of smaller distinct and disconnected ranges occupied from between one and 18 months (Stanley Price, 1988). Adult bulls followed a strategy of accompanying the breeding herds or ‘herd living’ and were not territorial.

DEVELOPING WILD POPULATION 19861989 In June 1986 the three-year drought was broken, supplementary feeding ceased and the two herds broke into sub-groups (Stanley Price, 1989). Later, a sterile but dominant bull was castrated and further immigrants were released. Rain fell each year from 1987 to 1989. These factors

were contributory to the more rapid growth (A = 1.35) in the wild population from 1986 to 1989 with A reaching 1.55 in 1988.

Immigrants After more than three years with no immigrants, 2.2 animals were brought from Jordan in 1987 and 3.3 from the USA in 1988. These animals were held together and released in October 1988. One F aged nine years proved to be too old for life in the wild and was transferred to the BBBC, where a programme of captive breeding with 99 not suitable for release and a single bull from Bahrain had been successful. In 1989 2.1 animals from BBHC and 3.3 from the USA were taken to the reintroduction area ready for releas(:. A hand-reared 6 and a 9 injured in captivity were unsuit- able for release and transferred to the BBBC. This fourth herd was released in September 1989.

Fecundity and mortality Fecundity was high and calf survival to 30 days was 96%. The average age at first calf during

Plate 2. Male and 0 Arabian oryx with calf.

84

this period was 25.33k3.46 months indi- cating that desert-born animals growing up with good grazing were able to conceive earlier than those born in the previous dry years and as early as captive animals with high quality diets. The oryx has a mean gestation period of 264 days (Stanley Price, 1989) and is capable of breeding all year round. Desert-born animals calving for the first time during this period and including all calves conceived before the drought of 1991 had a mean calving interval of 1 1.22 k 0.65 months indicating that most conceptions occurred at just less than three months post partum.

The number of breeding 88 increased as the first desert-born bulls sired off- spring although immigrants continued to hold breeding positions. As numbers of adult 86 increased the severity of fighting seemed also to increase and first of a series of fight-related deaths was recorded in 1989. Otherwise mortality was low with just six deaths in the wild recorded during this period.

Social and spatial organization After the break up of the herds a more dynamic social organization developed. With hier- archical dominance it was based on pre- dominantly 0 groups with following calves. Group size varied considerably and experienced frequent changes. Males tended to leave maternal groups at maturity and there was the first indication of territoriality as some desert-born breeding bulls established territories.

The pattern of range expansion, through response to rain events, recorded during the early years continued. As a consequence of the unpredictable nature of the rainfall the oryx, unlike many other ungulate species, do not occupy ranges for life but tend to occupy small, sometimes overlapping, areas each of which is part of a greater home range and which is re-visited and expanded as conditions dictate; as a result the home range conti- nues to expand throughout an animal’s life.

WILD POPULATION 199@1992 This period was marked by a slowing in growth rates to 1.17 1” as natural mortality factors came increasingly into operation and as the population faced drought; its first real test of fitness. Male territoriality continued to develop.

Immigrants The need for more immi- grant animals was recognized in Oman and shipments from zoos in the USA were scheduled for 1990 and 1991. As for all previous shipments immigrant animals were required to meet Oman’s stringent health requirements. However. in 1990 as a consequence of concern about existing techniques for testing for tuberculosis and the occurrence of the disease in the species in Arabia (Greth et al., 1992, see also this volume) regulations became so strict that the captive source in the USA was no longer able to comply (Woodford & Rossiter, 1993). The necessary testing was completed but isolation of animals, some of which tested positive to the serological test for tuberculosis, could not be arranged. This combined with the outbreak of war in the Gulf meant that no immigrants from outside Oman were received during this period.

This was a setback and occurred just before the reintroduction area was to experience drought in 1991. Late in that year with no prospect of further immi- grants, rising mortality in the wild and a population of just over 100, a number of animals which would have died in the wild were taken into captivity. This included the return of two 99, one with following calf, to the release enclosure and the hand-rearing of two calves abandoned by their dams. They were joined in the enclo- sure by further immigrants from the BBBC. The strategy proved successful with the combination of captive-bred, hand-reared and wild-born animals forming a cohesive herd which was released to the wild in October 1992, the fifth to be released. Fecundity and mortality Fecundity remained high in 1990 with animals born

UNGULATES 85

between 1987 and 1989 producing their first calves. There was one set of twins, the first record of twinning for the species in the wild. The first born, a 6, was dead and the second, a 0 , developed normally but died during the drought in early 1992. The number of bulls siring calves continued to increase during this period and immigrant animals were finding fewer breeding posi- tions. As the process of reintroduction continues it seems increasingly likely that immigrant bulls will have little breeding success. Older wild-born bulls were also holding fewer breeding positions as competition from younger animals increased and by October 1992 the mean age of 66 holding potential breeding posi- tions was 4.78 f 2.49 years.

Rates of mortality among calves and adults increased in 1990. Calf mortality to 30 days was 20% and causes of death included further cases of infanticide involving both true and surrogate sires. However, there was no indication of an increase in the frequency of infanticide which was first recorded in 1983 by Stanley Price (1989). The main cause of

death among adult animals was fighting with a number of bulls dying several days after receiving abdominal punctures or in one case immediately after a main blood vessel was ruptured b y a horn. Although fighting deaths have been recorded in cap- tivity these were the first records in the wild.

In 1991 drought conditions led to a rapid loss of condition in the animals in most age classes with lactating 99 among those deteriorating most quickly. Fecun- dity fell and mortality rose as the popula- tion experienced its first drought without supplementary feed. Mortality was highest among the youngest and oldest animals. First-year mortality for both sexes, including animals recovered for hand-rearing, exceeded 50%. The primary cause of death was malnutrition and dehydration.

The core of the population, however, survived the drought which was broken early in 1992, but subsequent flooding in April led to the death of a further three oryx. Fecundity remained low with few births for several months following the

Plate 3. Male Arabian oryx sparring.

86

rain suggesting that during the height of the drought there were no successful conceptions. However, in spite of their poor condition, animals conceived im- mediately after the drought was broken and 15 calves were born in December 1992 with others expected early in 1993; it is apparent that oryx, like other animals in an environment of unpredictable rain- fall, respond quickly to improved conditions.

By 3 1 December 1992 the wild popula- tion numbered 146 animals of which 87% was desert born. The sex ratio was almost 1 : l . The age structure shows a young population with a high reproductive capacity (Fig. 2). With further rain in October, when a tropical storm passed over the reintroduction area, a high rate of increase was predicted for 1993.

Social and spatial organization By September 1992 the area known to the oryx totalled 14 121 km2 with 18 breeding herds, mean size 5.8 f 2.94, occupying ranges with a mean size of 57.1 7 F 34.3 1 km2 and 7.3% of the known area. The herds comprised 84% of the population with the remaining 16?4 consisting of solitary bulls and one small bachelor herd.

While OF continued to expand range,

y e a r s

14

12

dd 9E Fig. 2. Age structure of the Arabian oryx population at 31 December 1992.

mature bulls increasingly adopted a terri- torial strategy as densities of Sr.; increased. The strategy of herd-living recorded during the early years was no longer seen. Other bulls were not terri- torial and some. immigrant animals in particular, roamed the reintroduction area. Observations suggested that these 8s were not able to hold territories.

Territories of ten desert-born dominant 83 holding and maintaining potential breeding position in 1991 are shown in Fig. 3. In some cases animals held two territories during the 12-month period. while in others the same territory was held for over three years. Territories ranged from 4 4 4 5 3 km2 with a mean overlap of 4%. An overlap of 21 % occurred between the ranges 005B and 002 and the two holders fought, with oryx 005 receiving fatal wounds. Except for oryx 00SA and 007A each animal was solitary within its range for 26f 12% of sightings.

The social and spatial organization that has developed in recent years is similar to that for other ungulates, particularly the Fringe-eared oryx 0. gazella cullotis as described by Wacher (1988). Not surpris- ingly, given the lower primary produc- tivity of the Arabian deserts, the major differences between the two species were the smaller herd size and greater home range of the Arabian oryx.

GENETIC CONSIDERATIONS The studbook shows the world popula- tion of oryx to be descended from 17 wild- caught founders in captivity in the USA and Arabia (Dolan, 1977 rf sey.). Breeding records indicated that at least two other founders existed in Arabian collections. Sixteen founders were repre- sented in the Oman wild populations although founder contributions were not equal (Fig. 4) with 93% of the genetic diversity coming from USA founder lineages. All but one of the release stock brought to Oman prior to 1987 were from the USA. Animals coming from Jordan in 1987 and offspring of a single animal from Bahrain held at the BBBC increased

UNGULATES

Fig, 3. Territories of desert-born dominant Arabian oryx d$ in 1991. Inser Location of the Jiddat al-Harasis in the Sultanate of Oman.

the numbers of founders represented. Founder representation remained skewed and three founder lineages occurring in other Arab collections were not repre- sented. A descendant population with unequal representation of founders will contain less genetic variability than one in which all founders have made equal contributions to future generation (Lacy, 1989).

GENEDROP analysiis (computer simu- lation developed by G. Mace, Zoological Society of London and modified by R. Lacy, Chicago Zoological Society) showed that the fraction of wild-gene diversity of each founder retained in the Oman wild population ranged between 22% and 89% with a mean of 57%. Mean retention among USA lineages fell slightly from 72& 11% in 1988 (Mace, 1989) to

88

'~ genetic representation inbreeding C O e f f

201

]

1 2 58

8 10 141 190 8000 140 189 809 9000

studbook no.

Fig. 4. Founder representation among Living Arabian oryx at 31 December 1992.

71 i 17% by 31 December 1992, indi- cating that the wild population has retained its broad genetic base although importation of more stock from Arabian collections is required.

With so few founder animals inbreeding has occurred in the Oman population (Fig. 5). A downward trend to 1986 was followed by an increase in 1987 coinciding with the first close sib matings as desert 63 reached maturity (Spalton, 1992). A decrease occurred after 1987 with new immigrants in 1988 and 1989 having an ameliorating effect on any increase in inbreeding. A slight increase occurred in 1991 and calves born in 1992 had an inbreeding coefficient of 0.1 26 & 0.08 1. A level of 0.125 is generally considered undesirable (Mace, 1989).

Inbreeding has been linked to juvenile mortality in many species including ungu- lates (Ballou & Ralls, 1982). Mace (1989) reported that for births prior to 1987 those calves not surviving to one year were significantly more inbred than those that survived. An examination of all births to December 1989 showed no such relationship (Spalton, 1992) and similarly for all births to December 1992 there was

0 . 3

0.2

0.1

f"' , I

I '

& ! - 1 (1)

T l I l l l l l I I i I I 1980 1982 1984 1986 1988 1990 1992

Fig. 5. Inbreeding coeffients ( & S.D.) of Arabian oryx calves of known paternity born each year from 1980-1992.

no significant difference in inbreeding coefficiencies between those surviving and those dying at less than one year (Mann- Whitney U = 1504, P= 0.096). Although inbreeding has increased, the evidence for any effect on calf survival is currently lacking. The trend of increasing levels of inbreeding is inevitable in a population with so few founders although the impor- tation of stock from under-represented lines can help to reduce this effect.

The effective population size (Ne, as defined by Wright (1931)) of the reintro- duced population was estimated to be 7.5 in 1986 (Spalton, 1992) and by 1992 had increased to 24.6. To achieve the desirable goal of an effective population of 50-100 (Mace, 1989), an actual population of between 264 and 528 is required.

Cribiu et al. (1990) described the occur- rence of a chromosomal translocation in captive Arabian oryx in Taif, Saudi Arabia. Studbook information indicated that the dam of a homozygous carrier of the Robertsonian translocation originated from a captive herd in Qatar. Subse- quently karyotyping of two animals arriving in Oman from Jordan showed

UNGULATES 89

them both to be translocation carriers. Work is continuing with investigations into the breeding success of carriers and occurrence of the translocation in the wild population (see also Greth et al., this volume).

FUTURE A computer simulation developed by J. Ballou and L. Bingman of the National Zoo, Washington, DC, was used to predict numbers in the wild during the next six years. Projections using fecundity and mortality data for the wild popula- tion since 1982 showed that numbers could reach 204 by the end of 1994, 282 by the end of 1996 and 388 by the end of 1998. Given the current favourable condi- tions the short-term projections are prob- ably an underestimate and the population could reach 200 by early 1994. Although no attempt was made to incorporate stochastic risks, the fecundity and mortality schedules included data collected during drought conditions.

Further importation of immigrants would help to accelerate the growth of the population in the wild. To enable the further regular importation of stock effec- tively screened for tuberculosis, a quaran- tine facility is to be built in Muscat. This will allow an increased rate of immigra- tion without compromising the apparent good health of the population by relaxa- tion of the current specific conditions for import permits and zoo-sanitary certifi- cates for wild ruminants imported into Oman.

It is difficult to predict carrying capacity in an environment where the pattern of rainfall, both temporally and spatially, is so uncertain. The reintroduc- tion area including adjacent sand dunes covers 34 000 km2 (Stanley Price, 1989). To date the oryx have utilized 41% of that area although they occupy only a small fraction at any time. The conse- quence of such a unpredictable rainfall pattern is an ever-fluctuating carrying capacity.

The carrying capacity of the reintroduc-

tion area will also depend on the numbers of other herbivores, particularly domestic stock, with which the oryx must compete for an often limited resource. Observa- tions suggest that during times not following widespread rainfall the major factor influencing the ranging behaviour of the oryx is the location of bedouin goat herds. Numbers of domestic stock, supported with supplements during drought, are increasing and will reduce the potential carrying capacity of the rein- troduction area. The Jiddat al-Harasis has been proposed as a World Natural Heri- tage Zone and, combined with regional development plans, associated legislation should ensure that the integrity of the oryx habitat is not lost

CONCLUSIONS Since 1986 the population has developed from two predominantly immigrant herds with strong site fidelity into a more dynamic population with a complex social and spatial organizaticm independent of the release area where group size varied considerably and frequently. Growth rates have slowed in recent years as natural mortality factors have come increasingly into operation but the popu- lation has continued to grow and expand its known range. Important changes in spatial organization ha.ve occurred with 33 becoming increasingly territorial. Desert-born animals appear to be fitter than immigrant animalis and the popula- tion has survived drought; the first real test of fitness since it became truly inde- pendent. Levels of inbreeding do appear to be increasing but milch of the genetic diversity of the founder stock has been maintained.

Provided the animals are not faced by severe drought or other stochastic events, a target population of at least 264 should be reached by 1996 or with further immi- grants by 1995. At least until that point, the level of effort and support of the first 12 years of the reintroduction needs to be maintained if the long-1:erm success is to be assured.

90

ACKNOWLEDGEMENTS The Arabian Oryx project, originally inspired and continuously supported by His Majesty Sultan Qaboos bin Said, is funded wholly by the Government of the Sultanate of Oman. I acknowledge the work of M. R. Stanley Price and T. H. Tear during the first nine years of the project. I thank R. H. Daly, Adviser for Conservation of the Environment, and E. R. L. Jones, Field Manager of the White Oryx Project, for their advice and support. I am most grateful to M. Fisher of the Sultan Qaboos University for constructive criticism made during the preparation of this manuscript.

REFERENCES BALLOU, J. b: RALLS, K. (1982): Inbreeding and juvenile mortality in small populations of ungulates: a detailed analysis. Biol. Conserv. 24: 239-272. CRIBIU, E. P., ASMODE, J.-F., DURAND, V., GRETH, A. b: ANAGARIYAH, S. (1990): Robertsonian chromosome polymorphism in the Arabian oryx (Oryx leucoryx). Cytogenet. CellGenet. 5 4 161-163. DOLAN, J. (1977 et seq.): The Arabian oryx studbook. San Diego: The Zoological Society of San Diego. GRETH, A,, SUNNUCKS, P., VASSART, M. & STANLEY, H. (1992): Genetic management of an Arabian oryx (Orq'x leucoryx) population without known pedigree. In Ungulates 91: 77-83. Spitz, F., Janeau, G., Gonzalez, G. & Aulagnier, S. (Eds). Paris: SFEPM and Toulouse: IRGM. FITTER, R. (1982): Arabian oryx returns to the wild. 0ry.u 1 6 406-410. LACY, R. C. (1989): Analysis of founder representation in pedigrees: founder equivalents and founder genome equivalents. Zoo B i d . 8: 11 1-123. MACE, G. M. (1989): Genetic status of Arabian oryx

in Oman. In Strategies and management Jur continuing reintroduction of the species to the u'ild. Oman: Office of the Adviser for Conservation of the Environment, Diwan of Royal Court. NOTTROTT. R. (1986): RANGE (Range Analysis Program) user's guide and manual. Oman: Office of the Adviser for Conservation of the Environment. Diwan of Royal Court. SPALTON, J . A. (1992): The Arabian orbx i0 ry .u leucoryx) reintroduction project in Oman: 10 years on. In Ungulates 91: 343-347. Spitz. F., Janeau, G. , Gonzalez, G. & Aulagnier, S. (Eds). Paris: SFEPM and Toulouse: IRGM. STANLEY PRICE, M. R. (1986): The reintroduction of the Arabian oryx Orjx 1eucorj.x into Oman. I n / . Zoo

STANLEY PRICE, M. R. (1988): Fields operation and research in Oman. In Conservation and biology of deserr antelopes: 18-34. Dixon, A. &Jones, D. (Eds). London: Christopher Helm. STANLEY PRICE, M. R. (1989): Animal re- introductions: the Arabian orj'.x in Oman. Cambridge: Cambridge University Press. WACHER, T. (1988): Social organisation and ranging behaviour in the Hippotraginae. In conservation and biology of desert antelopes: 102-113. Dixon. A. & Jones, D. (Eds). London: Christopher Helm. WOODFORD, M. H. & ROSSITER, P. B. (1993): Disease risks associated with wildlife translocation projects. In Health and management of free-ranging mammals. Part 2. O.I.E. Scientific and Technical Revien, 12(1): 1 15-1 35. WRIGHT, S. (1931): Evolution in Mendelian populations. Genetics 1 6 97- 159.

Yb. 24/25: 179-188.

Manuscript submitted 22 January 1993