REINTRODUCTION OF EXTINCT SPECIES OF LARGE HERBIVORES TO GILÉ NATIONAL RESERVE, ZAMBÉZIA PROVINCE, MOZAMBIQUE

February 2010

TITLE: REINTRODUCTION OF EXTINCT SPECIES OF LARGE HERBIVORES TO GILÉ NATIONAL RESERVE, ZAMBÉZIA PROVINCE, MOZAMBIQUE

AUTHOR: IGF FOUNDATION

PUBLICATION: February 2010, Musseia, Gilé National Reserve, Zambézia Province, Mozambique

FUNDING: FFEM & IGF FOUNDATION

NATURE OF THE STUDY: Appraisal survey

COUNTRY CONCERNED: Mozambique

KEY-WORDS: Reintroduction - Extinct species – Large herbivores – Capture – Translocation – Monitoring

ABSTRACT:

This study was carried out within the project for the co-management of the Gilé National Reserve, Mozambique under the auspices of the Ministry of Tourism of the Republic of Mozambique and the IGF Foundation. The study was funded by the FFEM, the IGF Foundation and private donors. The purpose of the study was to assess the feasibility of the reintroduction of extinct species of large herbivores to the Gilé National Reserve.

The Gilé National Reserve is considered as a suitable recipient area for succeeding a reintroduction operation given the recent and steady improvement of its management standards.

Four candidate species of large herbivores are selected for reintroduction, namely: - the Southern (Syncerus caffer caffer); - the Livingstone’s eland (Taurotragus oryx livingstonii); - the Nyassaland wildebeest (Connochaetes taurinus johnstoni); - the Crawshay’s zebra (Equus quagga crawshayi).

For a number of reasons, the ideal source area would be the Niassa National Reserve.

Risks are evaluated and alleviation means identified.

The implementation of the operation is phased: - pre-translocation phase: preparation of the recipient area and the operation; - translocation phase: capture, transport, holding and release; - post-translocation: monitoring.

The whole operation complies with the IUCN guidelines for translocation living organisms.

Cover picture: Gilé National Reserve (©IGF Foundation/Pascal Mésochina)

Reintroduction of extinct species of large herbivores in Gilé National Reserve 2 RESUMO

Este estudo foi realizado no âmbito do projecto para a co-gestão da Reserva Nacional do Gilé (RNG), Moçambique implementado consoante o Memorado de Entendimento entre o Ministério do turismo da República de Moçambique e a Fundação IGF. O estudo foi financiado pelo FFEM, a Fundação IGF e doadores privados. O objectivo deste estudo era avaliar a viabilidade da reintrodução de espécies extintas de herbívoros de grande porte na Reserva Nacional do Gilé.

A Reserva Nacional do Gilé é considerada como sendo um local de acolhimento conveniente para conseguir a operação de reintrodução dado o melhoramento recente e importante do seu maneio.

Quatro espécies potenciais de herbívoros de grande porte são seleccionadas para a reintrodução, nomeadamente: - o Búfalo da África austral (Syncerus caffer caffer); - o Elande de Livingstone (Taurotragus oryx livingstonii); - o Boi-cavalo do Nyassaland (Connochaetes taurinus johnstoni); - o Zebra de Crawshay (Equus quagga crawshayi).

Para múltiplas razões, a área fonte ideal seria a Reserva nacional do Niassa.

Os riscos são avaliados e as suas medidas de mitigação são identificadas.

A implementação da operação segue as fases seguintes: - fase de pre-translocação: preparação da área de acolhimento e da operação; - fase de translocação: captura, transporte, monitoria no boma e solta; - fase de post- translocação: monitoria.

A totalidade da operação segue os guiões da UICN relativos à translocação de organismos vivos.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 3 ACKNOWLEDGMENTS

The following persons must be acknowledged for their respective contributions:

- Cobus Raath, Wildlifevets.com, South Africa; - Mike Lagrange, African Wildlife Management and Conservation (AWMC), ; - Hans-Otto Reuter, African Wildlife Services (AWS), Namibia; - Philippe Chardonnet and Hubert Boulet, IGF Foundation.

Several other IGF Foundation collaborators took part in the feasibility study.

A number of other resource persons in Mozambique have been consulted on this matter.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 4 ACRONYMS AND DEFINITIONS

Camp: living and working place, in Gilé National Reserve, with only traditionally built houses (Nassere, Pipini, Naivocone, Muxayani)

Community fiscal: villager working with the anti-poaching teams and the administration of the GNR for the monitoring outside the Reserve

DNAC: Direcção Nacional das Áreas de Conservação, MITUR (National Directorate of Protected Areas)

DPT: Direcção Provincial do Turismo, MITUR (Provincial Service of Tourism)

FAO: Food and Agriculture Organisation

FFEM: Fonds Français pour l’Environnement Mondial (French GEF)

Fiscal: Portuguese name for a game scout, trained and assigned to a protected area for biodiversity monitoring and protection. The fiscal (fiscais in plural) is a Mozambican civil servant, managed by the Ministry of Tourism

Guard: camp or main camp warden employed by the Reserve

IFC: International Finance Corporation (Branch of the World Bank in charge of operations with the Private Sector)

IGF Foundation: International Foundation for the Management of Wildlife (Fondation IGF)

Main camp: living and working place, in Gilé National Reserve, with at least one concrete building (Musseia, Mulela, Lice, Nakololo, Namurrua, Etaga)

Miombo: Vegetation type characterized by evergreen trees belonging to the Legume family (Fabacae(Fabaceae), in particular Brachystegia, Julbernardia and/or Isoberlinia genus

MITUR: Ministério do Turismo (Ministry of Tourism)

MoU: Memorandum of Understanding

Ratoeira: gin trap used by poachers

Régulo: community authority officially recognized by the Mozambican state

RNG: Reserva Nacional do Gilé (Gilé National Reserve)

TA: Technical Assistant

ToRs: Terms of Reference

US$: United State Dollar

Reintroduction of extinct species of large herbivores in Gilé National Reserve 5 CONTENTS I. RATIONALE ...... 7 II. FEASIBILITY ...... 7 1. Suitability of the recipient area...... 7 1.1. Socio-economic suitability...... 9 1.2. Management suitability...... 9 1.3. Ecological suitability...... 9 2. Selection of the method ...... 13 3. Selection of candidate taxa for reintroduction ...... 14 3.1. Species ...... 14 3.2. Sub-species...... 15 4. Selection of individuals ...... 21 4.1. Buffalo ...... 21 4.2. Wildebeest...... 21 4.3. Eland ...... 22 4.4. Zebra ...... 22 4.5. Size and structure of founding populations...... 23 5. Selection of source populations ...... 23 5.1. General considerations ...... 23 5.2. Specific considerations...... 24 6. Selection of timing...... 27 7. Health implications...... 28 7.1. General implications ...... 28 7.2. Specific implications...... 29 8. Socio-economic implications...... 30 III. IMPLEMENTATION...... 30 1. Pre-translocation phase...... 30 1.1. Organisation ...... 30 1.2. Preparation ...... 31 2. Translocation phase ...... 31 2.1. The capture operation...... 31 2.2. The translocation operation...... 32 2.3. The holding and release operation...... 34 3. Post-translocation phase...... 37 3.1. Post-release monitoring...... 37 3.2. Long-term monitoring...... 37 IV. RISK ANALYSIS ...... 37 1. Risks associated with people ...... 38 1.1. Local communities ...... 38 1.2. Operators...... 38 2. Risks associated with animals...... 38 3. Risks associated with ecosystems...... 38 3.1. Source ecosystem...... 38 3.2. Sink ecosystem...... 38 V. COMPLIANCE WITH IUCN REINTRODUCTION GUIDELINES ...... 39 1. DEFINITION OF TERMS ...... 40 2. AIMS AND OBJECTIVES OF RE-INTRODUCTION...... 40 3. MULTIDISCIPLINARY APPROACH...... 41 4. PRE-PROJECT ACTIVITIES...... 41 4.1. Biological...... 41 4.2. Socio-economic and legal requirements...... 44 5. PLANNING, PREPARATION AND RELEASE STAGES ...... 45 6. POST-RELEASE ACTIVITIES...... 45 APPENDICES...... 47 REFERENCES...... 51

Reintroduction of extinct species of large herbivores in Gilé National Reserve 6 I. RATIONALE

In June 2007, the Honourable Fernando Sumbana, Minister of Tourism of the Republic of Mozambique and Dr. Philippe Chardonnet, Director of the International Foundation for the Conservation of Wildlife (IGF Foundation) signed an agreement for the long-term co- management of the “Reserva Nacional do Gilé” (RNG, Gilé National Reserve), Zambézia Province, Mozambique. The final goal of this program is the rehabilitation of the whole ecosystem of the Reserve in collaboration with local stakeholders. For reaching that goal, one of the main objectives is to restore the wildlife community which existed originally.

The aim of this study is to investigate the practicability of the reintroduction of major large mammal species to the Reserve (feasibility), to describe ways and means to succeed the operation (implementation), to assess eventual risks and identify risk alleviation methods (risk analysis) and to match the operation with the IUCN principles (compliance with IUCN guidelines).

The method used for conducting this feasibility study was a long-lasting brain storming exercise with a number of collaborators from various networks: wildlife authorities at central, provincial and local levels, professionals specialized in wildlife capture and translocation, wildlife management experts, IGF Foundation collaborators, etc, combined with a field mission to Gilé. A mission carried out for other purpose in Niassa national reserve, enabled to collect some useful data on the ecological context of this reserve.

II. FEASIBILITY

1. SUITABILITY OF THE RECIPIENT AREA

The recipient (or sink) area is the Gilé National Reserve. The Reserve is the only protected area of the Province of Zambézia. It extends on some 2,100 km² ( 210,000 ha) between Gilé and Pebane Districts in the north-eastern part of Zambézia Province and is comprised between 16°14’45” and 16°50’30’’ south and between 38°05’38’’ and 38°48°45’’ east (Maps 1 & 2). The following watercourses bound the RNG on its northern, eastern and western parts: Nanhope (northern sector), Molocué (eastern sector), Naivocone, Lice and Mulela (western sector). There is no geographic delimitation in the southern sector of the Reserve: the southern border is legally defined as a line parallel to the road that connects the Mualama and Nova Naburi localities (Map 2).

Reintroduction of extinct species of large herbivores in Gilé National Reserve 7

Map 1. Location of the Gilé NR

Map 2. Geographical limitations of the Gilé NR

Reintroduction of extinct species of large herbivores in Gilé National Reserve 8 1.1. Socio-economic suitability

Compared to other protected areas in Mozambique and the sub-region, the Gilé National Reserve is granted with an outstanding socio-economic environment:

• there are no human settlements and no human encroachments within the Reserve, except a few manioc fields restricted to the edge of the northern boundary; • the human population density is low around the Reserve, especially along the western boundary where there is no village at all; • the area is not very appropriate for both agriculture (low soil fertility) and livestock husbandry (occurrence of bovine trypanosomiasis); • there are some uses of natural resources within the Reserve, both wild flora and fauna, although the project is clearly improving the control and the management of these uses to make them more sustainable.

1.2. Management suitability

The National Reserve of Gilé is benefiting from the recent on-going project of long-term co- management by the Ministry of Tourism of the Republic of Mozambique and the International Foundation for the Conservation of Wildlife (IGF Foundation):

• the former management plan of the Reserve was just updated early 2010; • the global management standards of the Reserve are already steadily improving; • the infrastructures are under rehabilitation; • the surveillance (antipoaching, fire control, etc.) is strengthened thanks to a reinforced trained staff; • proper scientific exploration and monitoring have been launched; • a buffer zone is being officially gazetted; • the surrounding communities are consulted on a regular basis; • a rural development program is planned for improving the living standards in the buffer zone

1.3. Ecological suitability

The habitat is an important issue to be taken into consideration for the reintroduction of large mammals in GNR. The reason lies in the importance of matching habitats between the source site and the recipient site.

• Habitat suitability

Habitat similarity between source and sink areas is a key driver for the success of any reintroduction operation. Translocated animals have better chance to adapt themselves to the new habitat of their sink site when the habitat of their source site is similar. As a consequence, any reintroduction operation must investigate this similarity with particular emphasis on the sink/recipient area (Gilé National Reserve).

The Miombo landscape prevails in Gilé National Reserve. The Miombo deciduous woodland of the Reserve is very well preserved; it is a unique relict left over after human alteration of the Zambezian landscape.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 9 As defined by White (1983), the RNG falls within the Zambezian Regional Centre of Endemism phytogeographic unit, and within the Vegetation Type 26: ‘Dry Zambezian Miombo Woodland’. The landscape of the RNG consists of a miombo woodland / dambo grassland mosaic where deciduous miombo woodland is dominant in terms of cover percentage. The dambos are small to medium size (maximum 5 hectares) edaphic grasslands that are often flooded during the rainy season. The miombo woodland is heterogeneous and varies according to several biotic characteristics: height of the trees, shape of the tree canopy (vertical or horizontal), density of undergrowth (shrubs and herbs) and ratio between deciduous and evergreen tree species. The current project has conducted a vegetation survey for producing a vegetation map (Map 3; Prin, 2009).

The main candidate source area is the Niassa National Reserve where the Miombo landscape prevails in a very similar way as in the Gilé National Reserve.

• Water availability

Water sources are readily available. The RNG is drained by three major watercourses and by numerous small streams, some with permanent water and others simply seasonal. The three major watercourses are the Molocué river, which flows along the eastern boundary, the Mulela river, which flows along the south-western border and the Malema river, which flows in the core sector of the RNG. Other permanent watercourses have a much reduced flow during the dry season (e.g. Naivocone in the northern area, Muipige in the south-eastern sector and Nakololo in the core sector). There also several seasonal water-pools.

• Relief similarities

The landscape of the RNG is characterized by a gently sloping plain declining southwards from 200 to 100m above sea level, and by granite inselbergs emerging from the woodland. Overall is quite similar to the source area.

• Climate similarities

The climates in Gilé and Niassa are very similar. The Gilé climate lies within the Walter’s tropical summer-rainfall climatic zone (White, 1983), with a well-defined wet period between November and April and a dry period for the other six months (May-October). The annual average rainfall is around 800-1,000 mm. The temperature varies substantially during the dry season, from 13°C (mean minimum in June) to 35.7°C (mean maximum in October).

• The predator suitability

The populations of large predators in the Reserve [the (Panthera leo), the leopard (Panthera pardus), the spotted hyena (Crocuta crocuta) and the (Lycaon pictus)] are too small at this stage for threatening reintroduced animals of prey species..

Furthermore, since the four mentioned species of large predators present in Gilé NR are present in Niassa NR and even much more abundant there than in Gilé, the reintroduced animals are already well experienced in respect to anti-predation behaviour.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 10

Map 3. The Miombo landscape of the Gilé NR

Reintroduction of extinct species of large herbivores in Gilé National Reserve 11

• Suitability in respect to “escaping”

As formerly said, the Reserve is largely delimited by physical natural barriers all along the western and the eastern boundaries with large fast flowing rivers and rocks. The northern and southern limits are not as well marked. The northern border is close to agricultural fields with little chance for wild animals to be tempted for crossing human settlements. The southern boundary is a 5 km large buffer zone made of Miombo forest.

• Biodiversity richness

Biodiversity richness is an indicator of the fitness of the ecosystem and its ability to host reintroduced animals and facilitate their adaptation to their new environment. The biological communities are rich and diverse, reflecting a well structured trophic web (Carpaneto, 2001).

- Mammals’ presence:

The IGF Foundation and the DPT Zambezia have established a database of the wildlife species known to occur in the RNG. In its first stage, the database encompassed a number of observations carried out in RNG, particularly those published in former studies.

The first checklist of mammals in Gilé NR by Dutton et al. (1973) described 69 species and was then updated through various more or less intensive field surveys and/or ethnozoological interviews of local populations (Chande et al., 1997; Carpaneto, 2001; Gallego Lizon, 2002; Fusari & Cumbane, 2002; Chardonnet & Dobremez, 2005; Boulet & Lamarque, 2007) and through a thorough field survey of the RNG's large mammal species carried out between October and December 2007, by the IGF Foundation and the Direcção Provincial do Turismo Zambézia (Provincial Bureau of Tourism, Zambezia Province) (Mésochina et al., 2008).

Among the 69 species, 10 species were removed from the 2002 mammal checklist because they had been mistakenly reported for several reasons (see details in Mésochina, 2008). Fifty nine species of mammals were considered as occurring in the RNG according by the last effective survey (Carpaneto, 2001) and were reported as occurring in the management plan of the RNG (Fusari & Cumbane, 2002): 5 primates (3 monkeys and 2 galagos), 2 elephant shrews, 2 hares, 12 rodents, 17 carnivores, 19 ungulates, 1 pangolin and the antbear. It is worth quoting the Lord Derby’s anomalure (Anomalurus derbianus). Two species not previously listed (the dwarf mongoose [Helogale parvula] and a rodent yet to be determined) were observed by the 2007 IGF/DPT survey.

Among large herbivores present in Gilé NR, some of the most prominent are: elephant (Loxodonta africana), sable antelope (Hippotragus niger), greater kudu (Tragelaphus strepsiceros), waterbuck (Kobus ellipsiprymnus), bushbuck (Tragelaphus scriptus). Must be added: bushpig (Potamochoerus larvatus) and warthog (Phacochoerus africanus). The minimal population size of the elephant was estimated at 78 individuals by Mésochina et al. (2007) who described an apparent dynamic demography with a 32% immature ratio. Interestingly, the population seems to have increased since the last survey conducted in the Reserve which estimated the population to ca. 25 individuals (Martins & Ntumi, 2002), which is an encouraging indication of the improving management of the Reserve in recent years.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 12 Among large carnivores, the following species are present in Gilé NR: lion (Panthera leo), African wild dog (Lycaon pictus) and the spotted hyena (Crocuta crocuta), all three in extremely low numbers, leopard (Panthera pardus), side-striped jackal (Canis adustus), serval (Felis serval).

- Mammals extinctions:

The black rhinoceros (Diceros bicornis) and the Nyassaland wildebeest (Connochaetes taurinus johnstoni) still present in the early 1970s (Dutton et al., 1973), were already considered as surely extinct in 2002 (Fusari & Cumbane, 2002) and are thus possible candidates for reintroduction. The Crawshay’s zebra (Equus zebra crawshayi) has not been seen since 2002, when two local informants observed two individuals along Malema watercourse (Fusari & Cumbane, 2002); the 2007 IGF/DPT survey did not find any evidence of occurrence for the zebra which is thus considered as extinct. The same survey came up with serious doubts (very poor evidence) for the presence of two species, the southern African buffalo (Syncerus caffer caffer) and the Livingstone’s eland (Tragelaphus oryx livingstonii), making them candidates for the reintroduction programme.

The Lichtenstein hartebeest (Alcelaphus lichtensteini) had last been observed by a reliable informant who saw 2 individuals in September 2005 (Mésochina et al., 2007), but was not noticed by the 2007 IGF/DPT survey. It was thought extinct until mid-2009 when it was re- discovered in small numbers (1 individual and 5 individuals), thus excluding the species from the reintroduction program.

It is worth mentioning that the impala (Aepyceros melampus) has never been recorded in the region, thus excluding the species from an introduction program.

2. SELECTION OF THE METHOD

Several methods can be used to reach the goal of restoring the original herbivore community of the ecosystem.

• The ideal method is to protect the ecosystem to such an extent that it can reconstruct by itself naturally. This method relies on natural rebuilding of large mammal species by two ways: o Natural increase of the mammal species remaining within the Reserve: this process will happen anyway, whatever method is selected; o Natural reintroduction of the mammal species which disappeared from the Reserve: this process has the capacity to be effective if and only if the given species exists outside the Reserve, an hypothesis which is quite doubtful even though not impossible, especially for the large mammal species with very extensive home range such as elephant, wild dog, lion etc; however, the method remains extremely improbable for other mammal species.

• A faster method is to accelerate the natural process mentioned above by reintroducing the mammal species which have already disappeared from the Reserve; that is the method selected for the current operation.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 13 The reintroduction method has been selected as the method of choice for the following reasons:

• the reasons responsible for the extinction of the species are under control which is a prerequisite for the success of any reintroduction; • the reintroduction method does not preclude the other methods to restore wildlife, i.e. natural increase and natural reintroduction; • the reintroduction method is more efficient than other methods to rehabilitate extinct species.

The success of the reintroduction method relies very much on the following assumptions:

• A proper selection of the method used: the method must be closely driven by the IUCN reintroduction principles. • An improved management of the Reserve to optimize the short-term survival of the reintroduced animals and the long-tem development of their populations. • A proper selection of candidate species for reintroduction: the knowledge of the conservation status of wildlife in the Reserve helps to identify and prioritize the appropriate candidate species. • A relevant selection of source populations: issues such as taxonomy, genetics, health, access etc. must be investigated to identify the most appropriate source populations. • A high professionalism for conducting the operation all along the different phases: preparation, organisation, capture, translocation, release, monitoring. A reintroduction operation cannot be developed by a single person and needs the participation of several specialists.

3. SELECTION OF CANDIDATE TAXA FOR REINTRODUCTION

3.1. Species

Five large herbivores species are recognised as extinct species in the Reserve: - the black rhinoceros (Diceros bicornis) - the African buffalo (Syncerus caffer) - the eland (Taurotragus oryx) - the wildebeest (Connochaetes taurinus) - the zebra (Equus quagga)

All these species used to occur in large numbers some decades ago.

The black rhinoceros is not considered for reintroduction at this stage even though the Gilé National Reserve is supposed to be the last area having hosted the species in Mozambique. The reintroduction of the species will be studied at a later stage when the Reserve has reached a higher level of rehabilitation as a whole.

Four species are considered for introduction, namely buffalo, wildebeest, eland and zebra.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 14 3.2. Sub-species

Any reintroduction program of a given species must ascertain that the proper taxon is selected, to make sure that the subspecies reintroduced is the same than the one which disappeared from the area. In this respect, the four candidate species are examined for identifying the proper subspecies which occurred in the Reserve before becoming extinct.

• Buffalo

Class: MAMMALIA

Supercohort: LAURASIATHERIA

Cohort: FERUNGULATA

Superorder: CETARTIODACTYLA

Order: RUMINANTIA

Suborder: PECORA

Superfamily: Bovoidea

Family: Bovidae

Subfamily: Bovinae

Tribe: Bovini

Genus: Syncerus

Species: caffer

In the past, major variations in body appearance and skin colour led to the classification of 44 sub-species, although at present, only four are recognized

- the dwarf or red forest buffalo Syncerus caffer nanus, restricted to the swampy jungle and rain forests of West Africa stretching from Gambia to the Congo and to northern Angola - the north-eastern or Nile savannah buffalo S.c. aequinoctialis, from central East Africa, Chad, Sudan to Somalia and south to Tanzania - the north-western savannah buffalo S.c. brachyceros, from Senegal, through the Sahel to Chad - the southern or Cape buffalo S.c. caffer, occupying an area from southern Tanzania down to the Cape, including Namibia, Botswana, Zambia and Angola. This sub- species is the obvious choice for Gilé.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 15

Map 4: Distribution range of the African buffalo

• Wildebeest

Class: MAMMALIA

Supercohort: LAURASIATHERIA

Cohort: FERUNGULATA

Superorder: CETARTIODACTYLA

Order: RUMINANTIA

Suborder: PECORA

Superfamily: Bovoidea

Family: Bovidae

Subfamily: Antilopinae

Tribe: Alcelaphini

Genus: Connochaetes

Species: taurinus

Reintroduction of extinct species of large herbivores in Gilé National Reserve 16 Formerly distributed from southern Kenya southwards to northern and eastern Namibia, Botswana, the Orange River in South Africa, and Mozambique (East 1999; Estes in press), Common Wildebeest has also been introduced to regions outside its former distribution range, such as the Eastern Highlands of Zimbabwe and private farmland in Namibia (East 1999).

The ranges of the five subspecies are as follows (following Estes & East 1999 and Estes in press):

- C. t. taurinus (Blue Wildebeest). Namibia and South Africa to Mozambique north of the Orange River, and from Mozambique to Zambia south of the Zambezi River, and from south-west Zambia to south-east and southern Angola. - C. t. cooksoni (Cookson’s Wildebeest). Restricted to the Luangwa Valley, Zambia. May have ranged as vagrants onto the adjacent plateau into central Malawi. - C. t. johnstoni (Nyassa or Johnston’s Wildebeest). North of Zambezi River in Mozambique to east-central Tanzania, and formerly in southern Malawi, where now extinct. - C. t. albojubatus (Eastern White-bearded Wildebeest). Northern Tanzania to central Kenya just south of the Equator, west to the Gregory Rift Valley. - C. t. mearnsi (Western White-bearded Wildebeest of the Serengeti ecosystem). Northern Tanzania and southern Kenya west of the Gregory Rift Valley, reaching Lake Victoria at Speke Bay.

It is obvious that the sub-species that has to be reintroduced into Gilé is C. t. johnstoni (Nyassa or Johnston’s Wildebeest). This makes the sources fairly limited.

Map 5: Distribution range of the wildebeest

Reintroduction of extinct species of large herbivores in Gilé National Reserve 17 • Eland

Class: MAMMALIA

Supercohort: LAURASIATHERIA

Cohort: FERUNGULATA

Superorder: CETARTIODACTYLA

Order: RUMINANTIA

Suborder: PECORA

Superfamily: Bovoidea

Family: Bovidae

Subfamily: Bovinae

Tribe: Tragelaphini

Genus: Tragelaphus

Species: oryx

There are two eland species namely: - Tragelaphus oryx occurs in southern regions with three recognised sub-species: • Cape or southern eland Tragelaphus oryx oryx (Botswana, Namibia, southern Mozambique, southern Zimbabwe and South Africa); • Livingstone’s eland Tragelaphus oryx livingstonii (Angola, Zambia, southern Zaire, northern Zimbabwe, northern Mozambique and Malawi); • East African or Patterson eland Tragelaphus oryx pattersoni (Tanzania, Kenya, eastern Sudan, Ethiopia, Uganda and Rwanda); - Tragelaphus derbianus, the Giant or Lord Derby’s eland with two sub-species: • T. derbianus gigas in Cameroon, CAR, Chad, Nigeria, RDC, Sudan; • T.d. derbianus in Guinea, Mali, Senegal.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 18

Map 6: Distribution range of the eland

From this distribution, it is clear that the eland sub-species to be reintroduced to Gilé is Livingstone’s eland, Tragelaphus oryx livingstonii, which is the same eland that occurs in Niassa.

• Zebra

Class: MAMMALIA

Supercohort: LAURASIATHERIA

Cohort: FERUNGULATA

Superorder: PARAXONIA

Order: PERISSODACTYLA

Family: Equidae

Genus: Equus

Species: quagga

Reintroduction of extinct species of large herbivores in Gilé National Reserve 19 The six morphologically defined subspecies are distributed as follows (following Groves and Bell 2004):

- E. q. crawshayi (Crawshay’s Zebra) occurs in Zambia, east of the Luangwa River, Malawi, south-eastern Tanzania from Lake Rukwa east to Mahungoi, and Mozambique as far south as the Gorongoza District; - E. q. borensis ranges in north-west Kenya, from Guas Ngishu and Lake Baringo, to the Karamoja District of Uganda and south-east Sudan, east of the Nile River to the northern limit of the species at 32°N; - E. q. boehmi (Grant’s Zebra or Boehm's Zebra) is found in Zambia, west of the Luangwa River, west to Kariba, Shaba Province of DR Congo north to Kibanzao Plateau; Tanzania north from Nyangaui and Kibwezi into south-west Uganda, south- west Kenya as far as Sotik, and east Kenya, east of the Rift Valley, into southern Ethiopia and perhaps to the Juba River in Somalia; - E. q. chapmani (Chapman's Zebra) ranges from north-east South Africa, from about 24°S, 31°E, north to Zimbabwe, west into Botswana at about 19°S, 24°E, the Caprivi Strip in Namibia, and southern Angola; - E. q. burchellii (Burchell's Zebra) formerly occurred north of the Vaal/Orange Rivers, extending north-west via southern Botswana to Etosha National Park and the Kaokoveld, south-east to KwaZulu-Natal and Swaziland. It is now extinct in the middle of its range. E.b. antiquorum is now included in this subspecies; - E. q. quagga (Quagga) occurred in the former Cape Province, south of the Orange and Vaal Rivers and west of the Drakensberg. Now extinct.

Map 7: Distribution range of the zebra

It must be noted that although no DNA sampling has been done from the Zebra, the range of the sub-species E. q. crawshayi (Crawshay’s Zebra) should be move down to Gilé as its distribution goes as far down as Gorongosa, south of the Zambezi River.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 20

4. SELECTION OF INDIVIDUALS

4.1. Buffalo

• Social Considerations

Four social structures occur:

- Family groups: 6-15 individuals comprising an adult breeding bull, several adult cows, sub-adult heifers and calves of different ages; - Bachelor groups: 4-30 young bulls of 3-7 years, that migrate between family groups within the larger herd; - Adult bull herds: 4-12 mature bulls of >7 years that are not associated with a family group and keep to their own home range; - Outcast post-mature bulls: 1-6 individuals that keep to themselves in a small area and do not associate with other buffalo.

A definite order of dominance hierarchy exists in family groups. Within bachelor and bull herds a linear hierarchy exists according to age and body mass, with the oldest, heaviest animal being the dominant leader. During the rut the leading breeding bull forces the sub- adult bulls out of the family group to join bachelor herds. At an age of 7-8 years the bulls attempt to establish their own family groups or to replace a dominant bull from an existing family group. Defeated bulls stay within the bull herds until 15 years when they lose their hierarchal position and join cohesive, frustrated post-mature groups that are extremely dangerous and are known as "killer machines" or "dagga boys". Bulls and related bull herds account for 6-7% of the buffalo population.

• Selection of candidate individuals

As it is the intention to have maximum production from the buffalo relocated to Gilé, we suggest that only family groups with their breeding bulls be relocated. This results in a stable group structure at destination, and maximum growth potential. In years to come the young calves and old bulls will form the bachelor herds and outcast bulls, there is no need to relocate them. Relocating only breeding cows and bulls (no sub-adults) have worked in the past, but the adaptation at destination is certainly more difficult. Maintaining the population structure as close to normal as possible remains the ideal.

4.2. Wildebeest

• Social considerations

The blue wildebeest is a social animal. The structure of a population consists of:

- family groups of 10-150 individuals consisting of adult cows, heifers and calves of both sexes; - bachelor groups of 6-20 sub-adult bulls of 2-2.4 years age and non-active dominant adult bulls; - territorial breeding bulls which are dominant and mostly solitary.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 21 No hierarchy order of dominance exists amongst either cows in family groups or bulls in bachelor groups. Family bonding between cows and their offspring are weak and members of different family groups exchange frequently. Sub-adult bulls are chased out of the family groups by dominant bulls during the rut, and join bachelor groups. At 4.5 years age they leave the bachelor groups to establish their own territories. Some bulls remain with the bachelor herds and only break away during the rutting season for territorial behaviour. Other bulls maintain their territories for as long as environmental conditions remain favourable. Due to their migratory behaviour blue wildebeest do not have fixed home ranges. During good years they do not migrate but keep in a temporarily home range. Bachelor herds keep to the periphery of the roaming grounds of the family herds. When in transit the territories become floating islands that move along with the migration. The bull then defends a 30–50 m zone around himself and any cows associated with him. Territorial behaviour is mostly restricted only to the rutting period.

• Selection of candidate individuals

It is therefore suggested that family groups are relocated and if they include adult bulls, no further bulls be relocated. Relocating territorial breeding bulls have the risk that at destination they will move so far from the herds relocated that they will never benefit the population through breeding. If no bulls are in the breeding herds, then it will be more beneficial to relocate a small bachelor herd.

4.3. Eland

• Social considerations

The eland is a nomadic, migratory species that usually occurs in small groups. However, under favourable conditions such as on grass plains, very large herds of 300-1 000 animals occur consisting of family groups of 15-50 animals.

During the calving season the females form nursing groups that include the yearlings. After the season adult bulls rejoin the family herds and remain until the start of the next calving season when they break away. The bulls maintain a strict hierarchy of dominance while they are in the family herd. The sex ratio generally consists of one adult bull for every 8-12 mature cows. Young bulls of 2.5-5 years form bachelor groups of 5-10. When social maturity is reached they join family herds. Adult bulls form small herds of 3-6 outside of the mating season. Adult body size is reached at five years.

• Selection of candidate individuals

Ideally 2 or 3 family groups should be captured, including the large breeding bulls.

4.4. Zebra

There is an extremely important aspect of social behaviour in zebra that must be considered in any relocation program. One cannot mix family groups during capture or transport. This results in compartments not always being filled to maximum capacity during transport, and actual waste of space. However, mixing family groups, especially with such a long transportation, will be fatal. Similarly, at destination, provision must be made for each family group to keep them separately; otherwise fighting will erupt during confinement.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 22

4.5. Size and structure of founding populations

In any reintroduction program of a given taxon, the size of the founding population is important to identify in order to, make sure that the number of founders is sufficient for establishing a durable demographic development of the taxon in the long run. It has been decided to select cerca 30 individuals for each of the 4 candidate species.

Similarly, the structure of the founding population must be adequate for guaranteeing a sustainable reproduction of the taxon. The capture exercise will attempt to select already established breeding groups for maintaining the best possible social cohesion within the founding population of each candidate species. In an effort to balance the sex ratio, careful attention will be paid to make sure that enough males are present in breeding groups. If too few males are present in a breeding group, extra males will have to be added. Very old individuals will be tentatively avoided.

5. SELECTION OF SOURCE POPULATIONS

This section is examining the best possible sources to capture the founding animals to be reintroduced to the Gilé National Reserve.

5.1. .General considerations

As general rules, the source population must comply with the following requirements:

• Habitat:

Similar habitat in source and sink areas (see above).

• Genetic:

Proper taxon (see above).

• Predation:

If the recipient area hosts natural predators, the source population should not be naïve to them.

A wide range of large carnivore species occurs in Gilé NR. Nevertheless, since the predator population size is extremely small in Gilé NR at this stage, it does not threaten at short term the reintroduced animals which are all prey species. However, a natural increase in large predators is certainly expected in the coming years with the current improvement of the management standards. As a consequence, the source area must be populated with large carnivores.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 23 • Health:

Ideally, the health status in the source area must be sounder than -or at least equal to- the health status in the recipient area for avoiding the introduction of new diseases by the reintroduction program.

o The tuberculosis constraint:

- The tuberculosis issue is a key factor for identifying the source population since, once introduced in a given ecosystem, the pathogen is nearly impossible to eradicate as shown for example in Kruger NP; - Since tuberculosis is not known in Gilé NR, and since a lot of Protected Areas elsewhere are already infected with this disease, the clean (tuberculosis-free) health status of Gilé NR must be protected by all means from contamination by entering the pathogen via reintroduced animals; - All buffalo populations where tuberculosis is prevalent must be avoided as source of reintroduced animal for Gilé NR, which means that South Africa and Zimbabwe cannot be selected as source areas.

o The tsetse fly constraint:

- Since tsetse fly occurs in Gilé NR, the source populations must lie in a tsetse fly distribution area for preventing trypanosomiasis to affect the reintroduced animals; - This is another reason for avoiding South Africa as a source area.

• Access

The distance between the source area and the recipient area is a critical issue for lowering as much as possible the mortality rate during the translocation phase. Beside the distance, road conditions are also critical for the same reason.

Access within the source area should be reasonable for allowing decent conditions to the capture operation.

5.2. Specific considerations

• Buffalo

The southern or Cape buffalo occupies an area from southern Tanzania down to the Cape, including Namibia, Botswana, Zambia and Angola.

o Possible sources:

- Tanzania hosts by far the largest population of buffalo close to Mozambique with more than 100,000 individuals only in the neighbouring Selous GR; - South Africa and Zimbabwe produce surplus buffalo for translocation and sale; however, the tuberculosis constraint is even tougher for buffalo due to the high susceptibility of the species; as a consequence South Africa and Zimbabwe cannot be selected as source areas.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 24 - There are several possible source areas within Mozambique: Niassa National Reserve, Marromeu National Reserve, Quirimbas National park, several coutadas, game farms and open areas;

o Decision:

- Plan A: the source population should be a Mozambique tuberculosis-free population. To our knowledge so far, the buffaloes of the Niassa NR which have been tested are negative. For sanitary reasons (tuberculosis, trypanosomiasis etc.) as well as because of ecological and genetic proximity, the preference should be given to the Niassa NR; - Plan B: another similar source area within Mozambique, such as Quirimbas NP, coutadas, game farms, etc.

• Wildebeest

The Nyassa or Johnston’s Wildebeest occurs north of Zambezi River in Mozambique to south-east Tanzania, and formerly in southern Malawi were now extinct.

o Possible sources: Selous Game Reserve in Tanzania and Northern Mozambique.

o Decision:

- Plan A: Niassa NR; - Plan B: another area in Northern Mozambique (see Map 8).

Reintroduction of extinct species of large herbivores in Gilé National Reserve 25

Map 8: Distribution of wildebeest in 1960 in Mozambique according to Estes & East (2009)

• Eland

The Livingstone’s eland occupies an area in Angola, Zambia, southern DRC, northern Zimbabwe, northern Mozambique, southern Tanzania and Malawi.

o Possible sources:

- - Chimanimani National Park - Kazuma Pan National Park - Mashatu Game Reserve - - Kafue National Park (Zambia) - Etosha National Park (Namibia)

Reintroduction of extinct species of large herbivores in Gilé National Reserve 26 o Decision:

- Plan A: Niassa NR; - Plan B: another area of northern Mozambique such as Quirimbas NP, coutadas, game farms, Malawi (neighbouring Zambezia Province).

• Zebra

The Crawshay’s Zebra occurs in Zambia, east of the Luangwa River, Malawi, south-eastern Tanzania from Lake Rukwa east to Mahungoi, and Mozambique as far south as the Gorongoza District.

o Possible Sources: South Luangwa National Park in Zambia, Niassa NR. o Decision:

- Plan A: Niassa NR; - Plan B: another area in northern Mozambique such as Quirimbas NP, coutadas, game farms, etc.

6. SELECTION OF TIMING

The timing is a crucial factor of success with two considerations at stake:

• Seasonal temperatures

The temperature is critical for transporting wild animals as safely as possible, the lower the better. Given the temperature pattern in the area (see figure below), the best time of the year is end of June beginning of July.

Figure 1: Mean temperature patterns in Zambézia and Nampula Provinces

Reintroduction of extinct species of large herbivores in Gilé National Reserve 27

• The breeding season

- Buffalo

There are two peaks in calves births in buffalo in Southern Africa, one in October and a bigger one in March. Considering that it is not ideal to translocate very small calves or very heavily pregnant cows, the months of June to August will be the ideal time for relocation.

- Wildebeest

As their rutting season is from March to May, the ideal time to relocate will be there after, June to August, again, calves will be strong and the females will not be heavily pregnant.

- Eland

Breeding can be year round, but definite peak in the rain season. Calving is peaking from August, so much care will have to be taken during the capture operation to avoid heavily pregnant cows.

- Zebra

Breeding occurs year round but has a peak in the early rainy season. Care will be taken not to capture family groups with small foals, as they do not do well in long transportations.

7. HEALTH IMPLICATIONS

7.1. General implications

• Source area

In a reintroduction program in general, there is no health risk incurred by the source area.

As mentioned before, the health status in the source area must be sounder than -or at least equal to- the health status in the recipient are for avoiding the introduction of new diseases by the reintroduction program.

• Recipient area

Two issues must be considered:

- The sanitary status of the recipient area must be preserved from introducing new diseases, i.e. diseases that do not occur yet in the area. To our knowledge, there is no evidence for suspecting that any disease occurring in Niassa National Reserve does not occur in Gilé National Reserve. - Ideally, the reintroduced animals should not be exposed to diseases which do not occur in their original habitat of the source area, i.e. diseases for which they are naïve. Trypanosomiasis is not a matter of concern because tsetse flies occur in candidate source areas (they occur in the recipient area).

Reintroduction of extinct species of large herbivores in Gilé National Reserve 28

7.2. Specific implications

• Buffalo

Diseases to consider while working with and moving buffalo are corridor disease, Nagana fever or trypanosomiasis, foot-and-mouth disease and tuberculosis which can spread to cattle. Other diseases are anthrax, redwater disease, brucellosis, rabies, mange and rinderpest. Buffalo are not susceptible to heartwater.

Recent disease surveys in Niassa National Reserve showed that the buffalo population there:

- do carry all three strains of foot-and-mouth disease virus (FMD); - are positive for corridor disease; - are negative for brucellosis; - are negative for tuberculosis.

Although we have no information on the status of FMD in Zambezia Province, there is little chance that it differs from that of the northern Provinces. The decision lies with the veterinary officers of Zambézia Province.

• Wildebeest

Wildebeest are resistant to most tropical diseases except for the larvae of the nasal bot-flies (Kirkioestrus minentus), ocular-vascular myiasis caused by the larvae of the fly Gidoelstia sp. and malignant catarrhal fever. As the latter two diseases are highly contagious in domestic livestock, especially in cattle, wildebeest must be separated from them by a corridor of at least 1 km, which is not a matter of concern for Gilé NR.

• Eland

Eland are prone to parasites and diseases such as foot-and-mouth, tuberculosis, theileriosis, botulism, rinderpest and roundworms. They are intolerant of heavy tick infections. Eland are also susceptible to both redwater disease and heartwater carried by ticks. They cannot be translocated directly from a sub-tropical or bushveld region to a high rainfall, montane grassland habitat. When moving eland between different types of habitat, high mortalities can only be avoided by applying special acclimatization quarantine procedures.

The question of Eland to Gilé will revolve around the current tick load in Gilé. In Niassa most of the Reserve is burned every year, and during the buffalo disease survey, extremely low tick burdens were found. The same will possibly be true for the eland. Having low tick burdens in Niassa could mean that pre-immunity to tick born diseases have been lost. Thus, heartwater may result in high mortalities in Gilé, if it is present there since the whole Reserve burns every year as well. This point needs to be discussed with the local veterinary officers.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 29 • Zebra

Plains zebra are fairly resistant to most tropical diseases except for horse-sickness that can eliminate entire populations, if they had no previous exposure to it. Again, this topic has to be discussed with the veterinary services of both Provinces. They tolerate high infections of ticks and are not susceptive to foot-and-mouth disease or to malignant catarrhal fever.

8. SOCIO-ECONOMIC IMPLICATIONS

A reintroduction program may have socio-economic implications when problematic species are considered such as:

- dangerous species threatening human lives such as elephant or large predators; - crop-damaging species such as elephant again; - cattle-raiding species such as large predators.

The current operation is concerning species with no or very little risk for people, crops and livestock. The only species which might cause eventual problems is the buffalo which is a dangerous species, although the very low number of animals reintroduced and the very large surface of the Reserve prevent for many years the exposition of people and crops to eventual problem buffaloes.

Furthermore, in community-based programs such as the current Gilé project, local communities living in the buffer zone are participating to the project and well aware of its developments.

III. IMPLEMENTATION

1. PRE-TRANSLOCATION PHASE

1.1. Organisation

The way the operation will be conducted will very much depend on the source of the animals. However, for the purpose of this document we will work from the assumption that the animals will come from the Niassa Game Reserve in the North of Mozambique.

Obviously this will be a multi disciplinary approach and important role players will be:

- The relevant Central Government authorities; - The provincial directorates of Tourism (DPT) of the two Provinces; - The provincial directorates of Agriculture (DPA) of the two Provinces; - The administrators of the two Reserves and their staff; - The funding agencies; - The contracted capturers.

It is imperative that these bodies are all kept informed of the progress in the planning and that all have input at all levels.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 30

The decision which will have the most impact on the planning of the operation will be the veterinary requirements for the relocation, as this may request holding and testing at origin. This will bring another phase of temporary confinement at capture which has its own logistics. As stated earlier, the only disease that may necessitate this is foot-and-mouth disease, of which all buffaloes in Niassa are carriers and the status of Gilé is unknown.

It is further important to note that the numbers decided on can only be a guideline, and not a cut off point. The logistical costs of placing the capture team and infrastructures in Niassa are so high that optimal use must be made and the maximum number of animals transported to reduce the cost per capita. It must also be understood that in certain species, especially zebra, family groups cannot be mixed inside trucks, and this may result in all compartments not being filled to capacity. It may result in more trips than what is budgeted for, but this can easily be accommodated on the budget spreadsheet.

1.2. Preparation

• Administrative preparation

This will include: - Feasibility study - Agreements for source area, transportation and health issues - Selection and contracting of the operating team

• Technical preparation

This will include:

- Informing local stakeholders; - Preparing staff of the Gilé NR; - Building boma or camp in Gilé NR; - Securing an intensive protection zone (IPZ) around the boma or camp; - Identifying and preparing the itinerary (routes, diesel & water stops, etc.); - Improving access within the Gilé NR: enlarging and improving the access road for the trucks (bridges and other obstacles).

2. TRANSLOCATION PHASE

2.1. The capture operation

As far as possible the animals will be captured by means of mass capture, which is using plastic bomas and herding the animals in by helicopter. In the case of buffalo, this may be adapted to darting all animals once they are in the boma, as buffalo are very reluctant to climb into trucks. As animals cannot be herded for long distances to these capture sites (due to the possibility of capture myopathy), the rule remains that the "site is taken to the animals, and not the animals to the site”. This will result in building several of these capture sites. Even with the best of intentions, provision must be made that several animals will have to be darted from the air to fill trucks.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 31 All animals that receive radio collars for monitoring following release will also have to be darted, but this will be done once they are inside the bomas.

We foresee that the helicopter (or an ultra light aircraft if available) will find a suitable herd of a particular species for relocation. There after the ground team will move into the area and, considering many factors such as access, camouflage, wind direction, shade and many more, a capture site will be build over the next day or two. The animals will be herded into the site, which will have several compartments separated by curtains for splitting animals. This will allow us to dart the animals inside this boma that requires collaring.

The animals will be loaded through a crush, at which time they will all receive tranquilizers by means of pole syringes.

It must also be noted that in the case of eland and buffalo, large adult bulls will be darted in the boma and placed into a truck that has individual compartments for bulls, as they tend to fight in trucks irrespective of tranquilization, resulting in mortalities during transport. One truck will have removable “bull compartments” to ensure most effective loading and use of space inside the trucks.

If there is just little space left in the truck, it will be more cost effective to dart a few animals from the air to fill the truck, rather that to do a complete new herding drive.

Once a truck is full and the animals tranquilised, it will immediately take off and start the journey to the Gilé NR.

2.2. The translocation operation

• Constraints

Undoubtedly the transport is the most challenging aspect of the translocation process because of major logistic constraints comprising of:

- Distance to travel, estimated to be nearly 1,000 km: • Heat considerations; • Water, rest and food requirements; • Communication difficulties; • Therefore requiring: . Several 8 ton rigid trucks; . Accompanying light vehicles with necessary permits experience, knowledge of area and protocol; . Through pre survey of route to be taken. - The access roads within Gilé NR need to be improved since they are: • Sandy in nature; • Pot-holed; • Narrow; • Poor river crossings;

• Itinerary

The most suitable itinerary will have to be selected beforehand after investigating:

Reintroduction of extinct species of large herbivores in Gilé National Reserve 32

- the length in km and hours; - the surface of the roads (% in tarmac and dust track); - the relief (mountainous areas to cross); - the obstacles (large cities, bridges, slopes, etc.); - the availability of water for the animals and fuel for the trucks ; - other.

There are several possible itineraries between Mecula (Niassa NR) and Mulela (Gilé NR) (Table 1).

Table 1: 3 possible itineraries between Mecula in the Niassa NR and Mulela in the Gilé NR.

Itinerary 1Itinerary 2 Itinerary 3 Route Km Route Km Route Km Mecula - Marrupa 134 Mecula - Marrupa 134 Mecula - Marrupa 134 Marrupa - Maua 103 Marrupa - Maua 103 Marrupa - Maua 103 Maua - Cuamba 143 Maua - Cuamba 143 Maua - Cuamba 143 Cuamba - Mutuali 59 Cuamba - Mutuali 59 Cuamba - Mutuali 59 Mutuali - Malema 51 Mutuali - Malema 51 Mutuali - Lioma 49 Malema - Nauela 73 Malema - Lapia 75 Lioma - Gurue 35 Nauela - Alto Molocue 33 Lapia - Alto Molocue 91 Gurue - Socone 63 Alto Molocue - Mugeba 131 Alto Molocue - Mugeba 131 Socone - Errego 33 Errego - Nipiodi 33 Nipiodi - Mugeba 25 Mugeba - Mocuba 32 Mugeba - Mocuba 32 Mugeba - Mocuba 32 Mocuba - Maganja 96 Mocuba - Maganja 96 Mocuba - Maganja 96 Maganja - Mocubela 69 Maganja - Mocubela 69 Maganja - Mocubela 69 Mocubela - Mulela 91 Mocubela - Mulela 91 Mocubela - Mulela 91 Total 1015 Total 1075 Total 965

• Logistics

Special adaptations will have to be made to the vehicles, due to the long distances that will be travelled on dirt road. The dust being sucked into the compartments will certainly stress the animals and positive pressure will have to be created in the compartments to expel dust. Similarly, provision must be made to cool the animals down en route, should we experience a hot spell during the operation.

Diesel stops should be minimized, and care must be taken to ensure 24-hour availability of fuel in more remote areas, or diesel must be taken on the “escort vehicle”. Each truck should have at least two drivers, as turn around time on the trucks must be kept as short as possible.

Each truck load should have an “escort vehicle”. This vehicle could carry extra diesel, tools and manpower to assist in breakdowns, flat tires, etc. It also buys food and reduces stopping time tremendously. If a vet is on board, he can apply further tranquilization as required.

It is critical that communication is established between the truck and the persons receiving the animals in Gilé to ensure that they are expecting the truck, exactly when it will arrive, and that correct off loading procedures are followed, especially if off-loading is at night.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 33

2.3. The holding and release operation

Several options are discussed below, but the off-loading facilities will have to be erected in Gilé NR well beforehand, inspected and be ready to receive the animals after capture. Care should be taken here of access for trucks, crossings of rivers, turning of trucks at the ramp, ramp height, water for animals inside and in surrounding areas, shade and dense areas for hiding, and food for the animals (natural or supplemented).

Especially important are the off-loading facilities for buffalo. This will have to be re-enforced to prevent the animals from escaping, allowing time for regrouping, watering and feeding.

The receiving boma (Figures 2, 3 & 4) can be used for all species, and can afterwards be used as a “hospital” area for animals that require treatment. It must also be understood that this will probably be the first of many introductions to Gilé, and that it will be worth it to spend some funds now to always be ready to receive animals, should more become available.

Figure 2: Example of a receiving boma in Gorongosa NP

Reintroduction of extinct species of large herbivores in Gilé National Reserve 34 Release gate

wildebeest hartebeest 100x100m

Swivel gates buffalo eland Hinged on either side

waterbuck spare 70x70m

2m wide ramp Slide gate

Figure 3: Possible configuration of a boma

Fence only not covered

Figure 4: Possible boma with an exercise pen fence option

Several options may be considered for the construction of the boma: - Timber post made walls are more permanent and more look appealing should the boma become a tourist attraction; - Cheapest would be to use plastic tied to a fence backing, either diamond mesh or bonnox supported by poles 120mm butt at 2 metre intervals; - Second hand conveyor belting may be obtainable free from mining consortiums depending on level of contact which may be worth pursuing;

Reintroduction of extinct species of large herbivores in Gilé National Reserve 35 - Depending on taste and finances there are a number of other alternatives including 60mm butt poles placed adjacent to each other supported on a steel channel iron above the ground.

There are three different scenarios at off-loading and release, and we have to consider the advantages and disadvantages of them all. In order of preference they are:

• The incubator method (camp)

This is by far the best system in the long term for the re-introduction of species into a new large area such as Gilé NR.

An area or camp of about 700 to 1,200 ha is selected on the basis of access, central location, diversity of habitat, availability of water and easy monitoring of fences. It is fenced with game fencing and a comprehensive road network is maintained inside for monitoring. It has a centrally located off-loading ramp and small boma/enclosure attached to it.

The principle is simple. If 30 zebra are introduced into Gilé NR, monitoring will be extremely difficult: if they disperse, they will settle in different areas bringing very small gene pools to each area. If off-loaded in the incubator, they are allowed to breed, and once the population is firmly established, a certain number (family group) is released each year, but the breeding group is kept inside. Several species can be held at the same time.

The advantages are obvious. You will always have a breeding group, you can release into any area you want, the gene pool can be changed with new introductions, training of personnel is easier, multiple species, and smaller numbers to start with. It can also be used in the future as an IPZ (Intensive Protection Zone). Adult bulls in species like eland and buffalo can be re- introduced to the herds.

If a release of one family group in an area turns out to be a failure, you still have the breeding group vs point 2 and 3.

The disadvantage is obviously cost and time constraints, but the benefits far outweigh these. It is currently in use in Gorongosa, and has been used successfully in the re-establishment of populations in many national parks in Africa.

• Release after confinement at destination

A boma complex is erected at destination. This can be a permanent structure from poles or made from curtains of game capture plastic. The main purpose of these is to allow the animals to regroup (vs point 3). The animals are off-loaded into these and kept for a period of about 24 - 72 hours.

The animals are off-loaded into this boma, they are fed and monitored, and at any time passively released (the gate is opened and the animals are left to go out by themselves).

If the intention is to keep them longer, the structure will have to be more permanent. This has huge cost implications but very little benefit. There is no guarantee that the animals will stay in that area, in fact, the chances that they will move off is 99%, unless they are confined for months.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 36 The adult bulls that were transported separately cannot be placed into these together with the family groups; they will have to be released directly or kept in a separate boma.

The release gate must be huge (curtain) to maximize the chances of all animals leaving the boma simultaneously; otherwise the structure would have no purpose.

It is the second best option, but after release the animals may still split up, move huge distances and be difficult to monitor. It is however cheaper, faster and have been used successfully in the past.

• Release directly into the field

In this method of “hard release” the vehicle simply backs up to an off-loading ramp, the doors are opened and the animals allowed running out. This is suitable for smaller areas where the animals will find each other easily, but not recommended for Gilé NR. The problem here is that all animals may not leave the truck at the same time, and the herds get split up. Even though we think they will find each other, experience has proofed this is not the case in large areas.

3. POST-TRANSLOCATION PHASE

3.1. Post-release monitoring

An intensive monitoring program will be conducted just after the release and for 1 full year time. The protocol includes: - Ear-tagging of all the individuals; - Collaring of 5 individuals, preferably adult females, for each species (20 collars); - The collars will be simple VHF collars; - A special team will be hired for the task with a dedicated vehicle, VHF receivers and antennas and appropriate communication tools (satellite phone and VHF radio handsets).

3.2. Long-term monitoring

The management staff of Gilé NR will monitor the reintroduced animals on a long-term basis with the objective of measuring the progress of the animals in terms of: - range expansion; - population dynamic (breeding); - morbidity and mortality; - ecological adaptation; - interferences with human activities inside and outside the Reserve.

IV. RISK ANALYSIS

No reintroduction operation is devoid of any risk. The nature of the various risks must first be identified and their respective levels evaluated. Then all possible measures must be taken to alleviate the risks, bearing in mind that a zero risk level cannot be achieved.

Possible risks are assessed and risk mitigation measures identified.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 37

1. RISKS ASSOCIATED WITH PEOPLE

1.1. Local communities

Local communities may be affected (people, livestock, crops, water facilities etc.) by reintroduced animals when dangerous species are considered such as elephant or for example. It is not the case with the species concerned with the exception of the buffalo.

Even though possible accidents (people encounter) or incidents (crop damage) due to buffaloes may occur, they remain doubtful for the first few years after release due to the very low number of reintroduced animals, the large surface of the Reserve and the absence of resident population inside the Reserve.

1.2. Operators

Operators may potentially be subject to accidents during the capture and translocation phases due to the animals themselves or logistical incidents.

The highest possible professionalism in conducting the operation is the best mean to mitigate eventual unexpectations.

2. RISKS ASSOCIATED WITH ANIMALS

Animals may suffer from accidents during the capture and translocation phases of the operation. Long distance, bad roads, dust on the road and heat in the middle of the day are some of the main risk factors.

The highest possible professionalism in conducting the operation is the best mean to mitigate eventual problems. Again, despite all measures taken, a small mortality rate is usual during this kind of operation. It is expected that the overall mortality rate should remain below 10%.

3. RISKS ASSOCIATED WITH ECOSYSTEMS

3.1. Source ecosystem

In theory, a given source ecosystem may be affected by the capture of an excessive number of animals.

In the present case such a situation cannot occur since the respective population sizes of the four species considered in Niassa NR are considerably higher than the small number of animals to be caught.

3.2. Sink ecosystem

In theory, a given sink population could be negatively impacted if the reintroduced species:

Reintroduction of extinct species of large herbivores in Gilé National Reserve 38 - became invasive at the expense of the landscape or the existing wildlife: this scenario is not real; - were dangerous animals to people: it is not the case with a weak exception for buffalo which will be reintroduced in small numbers to a very large area without inhabitants; - were crop-damaging or cattle-raiding animals: similarly it is not the case; - were disease-carriers: the matter is discussed more extensively in former sections; - were genetically different to the formerly existent taxa: much attention has been paid to select the right taxa.

V. COMPLIANCE WITH IUCN REINTRODUCTION GUIDELINES

Any reintroduction operation must comply with the IUCN/SSC Guidelines for Re- Introductions, as prepared by the SSC Re-introduction Specialist Group and approved by the 41st Meeting of the IUCN Council, Gland Switzerland, May 1995.

The following exercise compares this planned re-introduction to their guidelines and highlights the areas of concern.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 39 The original text of the reference IUCN documents is reproduced in italic characters. The comment appears in blue characters.

1. DEFINITION OF TERMS

"Re-introduction": an attempt to establish a species (2) in an area which was once part of its historical range, but from which it has been extirpated or become extinct (3) ("Re- establishment" is a synonym, but implies that the re-introduction has been successful) . "Translocation": deliberate and mediated movement of wild individuals or populations from one part of their range to another.

"Re-inforcement/Supplementation": addition of individuals to an existing population of conspecifics.

"Conservation/Benign Introductions": an attempt to establish a species, for the purpose of conservation, outside its recorded distribution but within an appropriate habitat and ecogeographical area. This is a feasible conservation tool only when there is no remaining area left within a species' historic range.

Looking at these guidelines this operation will be a true re-introduction as the 4 candidate species used to exist in Gilé NR were they became locally extinct.

2. AIMS AND OBJECTIVES OF RE-INTRODUCTION a. Aims: The principle aim of any re-introduction should be to establish a viable, free-ranging population in the wild, of a species, subspecies or race, which has become globally or locally extinct, or extirpated, in the wild. It should be re-introduced within the species' former natural habitat and range and should require minimal long-term management. b. Objectives: The objectives of a re-introduction may include: to enhance the long-term survival of a species; to re-establish a keystone species (in the ecological or cultural sense) in an ecosystem; to maintain and/or restore natural biodiversity; to provide long-term economic benefits to the local and/or national economy; to promote conservation awareness; or a combination of these.

This project certainly falls within the aims and objectives of the IUCN.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 40 3. MULTIDISCIPLINARY APPROACH

A re-introduction requires a multidisciplinary approach involving a team of persons drawn from a variety of backgrounds. As well as government personnel, they may include persons from governmental natural resource management agencies; non-governmental organisations; funding bodies; universities; veterinary institutions; zoos (and private animal breeders) and/or botanic gardens, with a full range of suitable expertise. Team leaders should be responsible for coordination between the various bodies and provision should be made for publicity and public education about the project.

The planned operation is a multi-disciplinary exercise.

4. PRE-PROJECT ACTIVITIES

4.1. Biological

(i) Feasibility study and background research • An assessment should be made of the taxonomic status of individuals to be reintroduced. They should preferably be of the same subspecies or race as those which were extirpated, unless adequate numbers are not available. An investigation of historical information about the loss and fate of individuals from the re-introduction area, as well as molecular genetic studies, should be undertaken in case of doubt as to individuals' taxonomic status. A study of genetic variation within and between populations of this and related taxa can also be helpful. Special care is needed when the population has long been extinct.

In the case of this re-introduction, the sub-species selected are not questioned if compared to previous distribution data. The reasons for the loss are well known and all alleviation efforts are made by the Gilé Project. No genetic work has been conducted on the individuals that were in the Reserve.

• Detailed studies should be made of the status and biology of wild populations (if they exist) to determine the species' critical needs. For animals, this would include descriptions of habitat preferences, intraspecific variation and adaptations to local ecological conditions, social behaviour, group composition, home range size, shelter and food requirements, foraging and feeding behaviour, predators and diseases. For migratory species, studies should include the potential migratory areas.

The 4 candidate species have been extensively studied in the sub-region. In Niassa NR they are subject to a global aerial census every second year. Considering that the species planned to be introduced used to thrive in Gilé NR, calculated assumptions can be made that their critical needs can be met, especially considering the low numbers proposed.

• The species, if any, that has filled the void created by the loss of the species concerned, should be determined; an understanding of the effect the re-introduced species will have on the ecosystem is important for ascertaining the success of the re-introduced population.

There is no species that have filled the void in Gilé National Reserve.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 41 • The build-up of the released population should be modelled under various sets of conditions, in order to specify the optimal number and composition of individuals to be released per year and the numbers of years necessary to promote establishment of a viable population.

This point has been addressed in the discussions on viable populations.

• A Population and Habitat Viability Analysis will aid in identifying significant environmental and population variables and assessing their potential interactions, which would guide long- term population management.

No PHVA has been conducted. The issue of founding populations has been extensively discussed.

(ii) Previous Re-introductions • Thorough research into previous re-introductions of the same or similar species and wide- ranging contacts with persons having relevant expertise should be conducted prior to and while developing re-introduction protocol.

A number of experts have been consulted.

(iii) Choice of release site and type • Site should be within the historic range of the species. For an initial re-enforcement there should be few remnant wild individuals. For a re-introduction, there should be no remnant population to prevent disease spread, social disruption and introduction of alien genes. In some circumstances, a re-introduction or re-enforcement may have to be made into an area which is fenced or otherwise delimited, but it should be within the species' former natural habitat and range.

This criterion is met with the planned reintroduction.

• The re-introduction area should have assured long-term protection (whether formal or otherwise).

The area is a proclaimed protected area, a surveillance strategy is in place as well as a proper and updated management plan within the framework of the ongoing Gilé Project.

(iv) Evaluation of re-introduction site • Availability of suitable habitat: re-introductions should only take place where the habitat and landscape requirements of the species are satisfied, and likely to be sustained for the foreseeable future. The possibility of natural habitat change since extirpation must be considered. Likewise, a change in the legal/ political or cultural environment since species extirpation needs to be ascertained and evaluated as a possible constraint. The area should have sufficient carrying capacity to sustain growth of the re-introduced population and support a viable (self-sustaining) population in the long run. • Identification and elimination, or reduction to a sufficient level, of previous causes of decline: could include disease; over-hunting; over-collection; pollution; poisoning; competition with or predation by introduced species; habitat loss; adverse effects of earlier research or management programmes; competition with domestic livestock, which may be seasonal.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 42 Where the release site has undergone substantial degradation caused by human activity, a habitat restoration programme should be initiated before the reintroduction is carried out.

This has been done, and with the absence of humans in the Reserve boundaries, the area is pristine with no landscape degradation.

(v) Availability of suitable release stock • It is desirable that source animals come from wild populations. If there is a choice of wild populations to supply founder stock for translocation, the source population should ideally be closely related genetically to the original native stock and show similar ecological characteristics (morphology, physiology, behaviour, habitat preference) to the original sub- population.

The preferred source will be Niassa Game Reserve, which is the closest source and therefore within the sub species distribution range. It can be said that these populations were possibly fused in previous years.

• Removal of individuals for re-introduction must not endanger the captive stock population or the wild source population. Stock must be guaranteed available on a regular and predictable basis, meeting specifications of the project protocol. • Individuals should only be removed from a wild population after the effects of translocation on the donor population have been assessed, and after it is guaranteed that these effects will not be negative.

Once the source has been identified for each species, this exercise will be conducted, certainly by the supplier.

• If captive or artificially propagated stock is to be used, it must be from a population which has been soundly managed both demographically and genetically, according to the principles of contemporary conservation biology. • Re-introductions should not be carried out merely because captive stocks exist, nor solely as a means of disposing of surplus stock.

It is not the intention to use any captive stock, but only source animals from wild populations.

• Prospective release stock, including stock that is a gift between governments, must be subjected to a thorough veterinary screening process before shipment from original source. Any animals found to be infected or which test positive for non-endemic or contagious pathogens with a potential impact on population levels, must be removed from the consignment, and the uninfected, negative remainder must be placed in strict quarantine for a suitable period before retest. If clear after retesting, the animals may be placed for shipment. • Since infection with serious disease can be acquired during shipment, especially if this is intercontinental, great care must be taken to minimize this risk. • Stock must meet all health regulations prescribed by the veterinary authorities of the recipient country and adequate provisions must be made for quarantine if necessary.

As part of the “Multi Disciplinary Approach” mentioned earlier, the veterinary authorities of both Provinces/areas will be fully informed and incorporated into the team and a protocol will be established for the relocation, should any disease testing be necessary.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 43 These tests, their results and the delays they may cause, or confinement requirements they may imply, will be addressed once sources has been identified.

(vi) Release of captive stock • Most species of mammals and birds rely heavily on individual experience and learning as juveniles for their survival; they should be given the opportunity to acquire the necessary information to enable survival in the wild, through training in their captive environment; a captive bred individual's probability of survival should approximate that of a wild counterpart. • Care should be taken to ensure that potentially dangerous captive bred animals (such as large carnivores or primates) are not so confident in the presence of humans that they might be a danger to local inhabitants and/or their livestock.

It is the intention to relocate family units from the wild with no captive bred stock. Predators are not part of this operation.

4.2. Socio-economic and legal requirements

• Re-introductions are generally long-term projects that require the commitment of long term financial and political support. • Socio-economic studies should be made to assess impacts, costs and benefits of the reintroduction programme to local human populations. • A thorough assessment of attitudes of local people to the proposed project is necessary to ensure long term protection of the re-introduced population, especially if the cause of species' decline was due to human factors (e.g. over-hunting, over-collection, loss or alteration of habitat). The programme should be fully understood, accepted and supported by local communities. • Where the security of the re-introduced population is at risk from human activities, measures should be taken to minimise these in the re-introduction area. If these measures are inadequate, the re-introduction should be abandoned or alternative release areas sought. • The policy of the country to re-introductions and to the species concerned should be assessed. This might include checking existing provincial, national and international legislation and regulations, and provision of new measures and required permits as necessary. • Re-introduction must take place with the full permission and involvement of all relevant government agencies of the recipient or host country. This is particularly important in reintroductions in border areas, or involving more than one state or when a re-introduced population can expand into other states, Provinces or territories. • If the species poses potential risk to life or property, these risks should be minimised and adequate provision made for compensation where necessary; where all other solutions fail, removal or destruction of the released individual should be considered. In the case of migratory/mobile species, provisions should be made for crossing of international/state boundaries.

The socio-economic implications have been discussed. Local communities are part of the current Gilé Project. 3 of the 4 species are of no danger at all. The fourth species, buffalo, is well known to local inhabitants for having formerly thrived in large numbers in the recipient area. Clear communication to the local communities will be maintained.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 44 5. PLANNING, PREPARATION AND RELEASE STAGES

• Approval of relevant government agencies and land owners, and coordination with national and international conservation organizations. • Construction of a multidisciplinary team with access to expert technical advice for all phases of the programme. • Identification of short- and long-term success indicators and prediction of programme duration, in context of agreed aims and objectives. • Securing adequate funding for all programme phases. • Design of pre- and post- release monitoring programme so that each re-introduction is a carefully designed experiment, with the capability to test methodology with scientifically collected data. Monitoring the health of individuals, as well as the survival, is important; intervention may be necessary if the situation proves unforeseeably favourable. • Appropriate health and genetic screening of release stock, including stock that is a gift between governments. Health screening of closely related species in the re-introduction area. • If release stock is wild-caught, care must be taken to ensure that: a) the stock is free from infectious or contagious pathogens and parasites before shipment and b) the stock will not be exposed to vectors of disease agents which may be present at the release site (and absent at the source site) and to which it may have no acquired immunity. • If vaccination prior to release, against local endemic or epidemic diseases of wild stock or domestic livestock at the release site, is deemed appropriate, this must be carried out during the "Preparation Stage" so as to allow sufficient time for the development of the required immunity. • Appropriate veterinary or horticultural measures as required to ensure health of released stock throughout the programme. This is to include adequate quarantine arrangements, especially where founder stock travels far or crosses international boundaries to the release site. • Development of transport plans for delivery of stock to the country and site of reintroduction, with special emphasis on ways to minimize stress on the individuals during transport. • Determination of release strategy (acclimatization of release stock to release area; behavioural training - including hunting and feeding; group composition, number, release patterns and techniques; timing). • Establishment of policies on interventions (see below). • Development of conservation education for long-term support; professional training of individuals involved in the long-term programme; public relations through the mass media and in local community; involvement where possible of local people in the programme. • The welfare of animals for release is of paramount concern through all these stages.

These points are covered in the management plan of the Reserve, but the reference to disease screening is relevant. This has been addressed in the section on “Possible Disease Implications”.

6. POST-RELEASE ACTIVITIES

• Post release monitoring is required of all (or sample of) individuals. This most vital aspect may be by direct (e.g. tagging, telemetry) or indirect (e.g. spoor, informants) methods as suitable. • Demographic, ecological and behavioural studies of released stock must be undertaken.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 45 • Study of processes of long-term adaptation by individuals and the population. • Collection and investigation of mortalities. • Interventions (e.g. supplemental feeding; veterinary aid; horticultural aid) when necessary. • Decisions for revision, rescheduling, or discontinuation of programme where necessary. • Habitat protection or restoration to continue where necessary. • Continuing public relations activities, including education and mass media coverage. • Evaluation of cost-effectiveness and success of re- introduction techniques. • Regular publications in scientific and popular literature.

The collaring of the animals is proposed, and the post release monitoring program has been established.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 46 APPENDICES

Buffalo Information Table Characteristic Bull Cow Adult body weight Kg 650-850 520-636 Adult shoulder height Cm 145-165 135-155 Expected longevity Years 24 20 Age of sexual maturity Years 2.5-3 3-3.5 Age of social adulthood (1st Years 7-8 4 mating) Gestation Months 11 1st Calf born at Years 5 Calving interval Years 20-26 Post maturity age (last mating) Years 15 18 Rutting season Mar-May (Lowveld Region) Calving season Oct-Apr Weaning age Months 5-7 Gender ratio: Natural (all ages) 1 1.8-2.4 Gender ratio: Production (all ages) 1 4 Mating ratio: Natural (adults) 1 3-6 Mating ratio: Production (adults) 1 12 Re-establishment: Absolute minimum 3 5 number needed Re-establishment: smallest viable 3 8 population size Spatial behaviour: home range Ha 100-400 5000-40000 Spatial behaviour: territory range Ha None None Large stock grazing unit (adult) LSU 1.4 per animal (78% of 1.1 per animal (78% of Dietary ratio (grass) diet) diet) Browsing unit (adult) BU 2.68 per animal (22% of 2.68 per animal (22% Dietary ratio (browse) diet) of diet) Maximum stocking load 10 animals per 1000 ha (at 400mm annual rainfall) 30 animals per 1000 ha (at 700mm annual rainfall) Minimum habitat size required ha 1200 Annual population growth 6-18% (mean 16%) Optimal annual rainfall 400-600mm Optimal vegetation structure: Grass height: 12-130 cm Woody canopy cover: 20-75%

Reintroduction of extinct species of large herbivores in Gilé National Reserve 47 Wildebeest information table Characteristic Bull Cow Adult body weight kg 210-260 170-200 Adult shoulder height cm 130-150 122-135 Expected longevity years 16 18 Age of sexual maturity months 36 18-27 Age of social adulthood (1st mating) months 4.5-5.5 2.5 Gestation days 255 1st calf born at years 3.5 Furrow interval months 10-18 Post maturity age (last mating) years - - Rutting season Mar-May & Aug (bi-modal) Calving season: Nov-Jan & May Weaning age months 8 Gender ratio: natural (all ages) 1 1.2 Gender ratio: production (all ages) 1 4 Mating ratio: natural (adults) 1 2-3 Mating ratio: production (adults) 1 6-10 Re-establishment: absolute minimum number needed 2 3 Re-establishment: smallest viable population size 3 8 Spatial behaviour: home range ha 600-1800 1000-2500 Spatial behaviour: territory range ha 0.5-1.5 (non None static) Large stock grazing unit (adult): LSU 0.52 per 0.52 per animal Dietary ratio (grass): animal (95% of diet) (95% of diet) Browsing unit (adult): BU 1.21 per 1.21 per animal Dietary ratio: (browse): animal (5% of diet) (5% of diet) Maximum stocking load 76 animals per 1000 ha (at 300-400 mm annual rainfall) Minimum habitat size required ha 300 Annual population growth 28-33% (mean 30%) Optimal annual rainfall 300-450 mm Optimal vegetation structure: Grass height: 2-12 cm Woody canopy cover: 20-45%

Reintroduction of extinct species of large herbivores in Gilé National Reserve 48 Eland information table Characteristic Bull Cow Adult body weight kg 650-910 400-600 Adult shoulder height cm 170 150 Expected longevity years 18-24 15-18 Age of sexual maturity months 18-24 15-18 Age of 1st mating years 5-7 1.8-2.5 Gestation days 271-279 1st calf born at years 2.5-3.3 Calving Interval months 10-12 Post maturity age (last mating) years - - Rutting season Year round (Peak Nov-Jan) Calving season Year round (Peak Aug- Nov) Weaning age months 5-6 Gender ratio: Natural (all ages) 1 1.5 Gender ratio: Production (all ages) 1 3.5 Mating ratio: Natural (adults) 1 8-12 Mating ratio: Production (adults) 1 10-15 Lamb/Calf birth ratio 1 1 Re-establishment: Absolute minimum 1 2 number needed Re-establishment: Smallest viable 3 7 population size Spatial behaviour: Home range ha Unlimited (>20000) Unlimited (>20000) Spatial behaviour: Territory ha None None range Large stock grazing unit (adult): LSU 0.9-2.0 per animal 0.09 per animal Dietary ratio (grass): (50-70% of diet) (50-70% of diet) Browsing unit (adult): BU 2.1-4.6 per animal (30- 2.3 per animal (30-50% Dietary ratio: (browse): 50% of diet) of diet) Maximum stocking load 45 Animals per 1000 ha (at 450-550 mm annual rainfall) Minimum habitat size required ha 3000 (500 with food supplement) Annual population growth 11-38% (mean 20%)

Reintroduction of extinct species of large herbivores in Gilé National Reserve 49 information table Characteristic Stallion Mare Adult body weight: kg 290-340 290-340 Adult shoulder height: cm 130-138 130-138 Expected longevity years 22 22 Age of sexual maturity months 36 16-20 Age of social adulthood (1st mating) years 5 2-2.5 Gestation months 12.5 1st foal born at years 3 Foal interval months 16 Post maturity age (last mating) years 12 17 Rutting season Year round Foaling season: Year round Weaning age months 10 Gender ratio: natural (all ages) 1 1.5 Gender ratio: production (all ages) 1 4 Mating ratio: natural (adults) 1 4 Mating ratio: production (adults) 1 6 Re-establishment: Absolute minimum number needed 1 3 Re-establishment: Smallest viable population size 3 5 Spatial behaviour: home range ha 10000-26000 7000-26000 Spatial behaviour: territory range ha None None Large stock grazing unit (adult): LSU 0.7 Per animal 0.7 per animal Dietary ratio (grass): (95% of diet) (95% of diet) Browsing unit (adult): BU 2.2 per animal 2.2 per animal Dietary ratio: (browse): (5% of diet) (5% of diet) Maximum stocking load 40 animals per 1000 ha (at 450-550 mm annual rainfall) Minimum habitat size required ha 800 Annual population growth: 15-29% Optimal annual rainfall: 450-550 mm Optimal vegetation structure: Grass height: 6-45 cm Woody canopy cover: 20-55%

Reintroduction of extinct species of large herbivores in Gilé National Reserve 50 REFERENCES

BOULET, H., & LAMARQUE, F. (2007). Co-gestion de la Réserve Nationale de Gilé, mission de préparation. Rapport Fondation IGF, 70 p. CARPANETO, G.M. (2001). The sustainable exploitation of wildlife in the National Reserve of Gilé (« Reserva do Gilé », Zambézia, Mozambique. Final report. Movimondo, 59 p. CHANDE, B., ZOLHO, R., CHIDIAMASSAMBA, C., LONGAMANE, F., & CUMBANE, R. (1997). Reconhecimento bio-ecológico da Reserva do Gilé. Ministério para a Coordenação da Acção Ambiental – Direcção Nacional de Gestão dos Recursos Naturais, 29 p. CHARDONNET, P., & DOBREMEZ, J.F. (2005). Projet Gilé du 10 au 20 août 2005 – Rapport provisoire. Fondation IGF, Paris, 20 p. DUTTON, T.P., DUTTON, E.A.R., & BALSINHAS, A. (1973). Preliminary ecological reconnaisance of the "Reserva especial do Gilé" and the adjoining "Regime de Vigilância", District of Zambezia. Instituto de Investigação Agronómica de Moçambique, Lourenço Marques, 40 p. ESTES, R. & EAST, R. (2009). The Status of the Wildebeest in the wild 1967-2005. Wildlife Conservation Society. Working Paper n°37. July 2009. FUSARI, A., & CUMBANE, R.J. (2002). The Game Reserve of Gilé management plan. Direcção Nacional de Áreas de Conservação, Mozambique, 136p. FUSARI, A. & CARPANETO, G.M. (2006). Subsistence hunting and conservation issues in the Game Reserve of Gilé, Mozambique. Biodiversity and Conservation, 15, 2477-2495. GALLEGO LIZON, J. (2002). Caterpillars, traps and bees. Livelihood Dependence on Wild Resources in the Natural Reserve of Gilé. Movimondo, 59 p. GROOVES, C. P. & BELL, C. H. (2004). New investigations on the taxonomy of the zebras genus Equus, subgenus Hippotigris. Mammalian Biology. 69: 182-196. INIA (1994). Carta de Solos: Província da Zambézia. Instituto Nacional de Investigação Agronómica, Maputo. LAGRANGE, M. (2009). Feasibility of reintroduction of Large Herbivores to Gilé National Reserve. Consultancy report, IGF Foundation, 5 p MARTINS, A., & NTUMI, C. (2002). Plant Diversity, Vegetation Ecological Status, Spatial and Temporal Movements of Elephants and Economic Value of Natural Resources in the National Reserve of Gilé, Zambézia Province-Mozambique. PRPGRG. Movimondo, Rome. MESOCHINA, P., LANGA, F. & CHARDONNET, P (2008). Preliminary survey of large herbivores in Gilé National Reserve. IGF Foundation and DPT Zambézia. MICOA (1999). Reconhecimento Florestal da Reserva de Caça do Gilé. Ministério para a Coordenação da Acção Ambiental-MICOA. Maputo. PRIN, T. (2009). Carte de végétation de la Réserve nationale de Gilé, Mozambique. Mémoire de Master, UNiversité de Montpellier. Fondation IGF & DPT Zambézia. RAATH, J.P. (2010). Desktop study on the reintroduction of four species of fauna to the Gilé national park in Mozambique. Consultancy report, IGF Foundation, 27 p. REPÚBLICA DE MOÇAMBIQUE – MINISTÉRIO DO TURISMO (2004). Plano de maneio 2003- 2007 Reserva do Gilé, 132 p. SMITHERS, R. H. N., & TELLO, J. L. P. (1976). Checklist and Atlas of the mammals of Mozambique. Museum Memoir No 8. Trustees of National Museum and Monuments of Rhodesia, Salisbury, 183 pp. WHITE, F. (1983). The vegetation of Africa. UNESCO, Paris.

Reintroduction of extinct species of large herbivores in Gilé National Reserve 51