Pachyderm 38 - January - June 2005
Transcription
Pachyderm 38 - January - June 2005
January – June 2005 Number 38 ISSN 1026 2881 IUCN journal of the African Elephant, African Rhino and Asian Rhino Specialist Groups The World Conservation Union January–June 2004 1 Chair reports / Rapports des Présidents 1 African Elephant Specialist Group / Groupe Spécialiste des Eléphants d’Afrique Holly T. Dublin 11 African Rhino Specialist Group / Groupe Spécialiste des Rhinos d’Afrique Martin Brooks 16 Asian Rhino Specialist Group / Groupe Spécialiste des Rhinos d’Asie Mohd Khan bin Momin Khan with/avec Thomas J. Foose and/et Nico van Strien 19 19 Research Changes in elephant numbers in major savanna populations in eastern and southern Africa J. Julian Blanc, Richard F.W. Barnes, G. Colin Craig, Iain Douglas-Hamilton, Holly T. Dublin, John A. Hart and Chris R. Thouless 29 The value of systematic recording of human–elephant conflict: a case study in south-eastern Tanzania Cyprian Malima, Richard Hoare and Julian Blanc 39 Elephant crop damage in the Red Volta Valley, north-eastern Ghana Patrick Adjewodah, Paul Beier, Moses K. Sam and John J. Mason 49 Nature and extent of human–elephant conflict in Bia Conservation Area, Ghana Moses Kofi Sam, Emmanuel Danquah, Samuel K. Oppong and Enoch A. Ashie SPECIES SURVIVAL COMMISSION Editor Helen van Houten Assistant Editor Dali Pam Mwagore Editorial Board Holly Dublin Esmond Martin Leo Niskanen Robert Olivier Nico van Strien Lucy Vigne Design and layout Damary Odanga Graphics Phillip Miyare Address all correspondence, including enquiries about subscription, to The Editor, Pachyderm PO Box 68200, 00200 Nairobi, Kenya tel: +254 20 576461 fax: +254 20 570385 email:[email protected] Web site: www.iucn.org/afesg This document has been produced with the financial assistance of the US Fish and Wildlife Service. The views expressed herein are those of the authors and can therefore in no way be taken to reflect the official opinion of the US Fish and Wildlife Service. No. 38 Cover: Elephant coming to water at Lake Banzena at the height of the dry season 2004, Gourma, Mali. Photo: Carlton Ward Jr. / www.LINC.us journal of the African Elephant, African Rhino and January–June 2005 No. 38 Asian Rhino Specialist Groups 59 Population survey of elephants in Okwangwo Division, Cross River National Park, Nigeria Emmanuel Obot, Clement Edet, Gabriel Ogar and Joy Ayuk 64 Elephants in Cote d’Ivoire—a warning for West African conservation Frauke Fischer 76 Analyse biométrie des pointes d’éléphants saisies dans le cadre de la lutte antibraconnage par les services de la conservation dans le massif du sud-est Cameroun Paul Noupa 82 Threats to the greater one-horned rhino and its habitat, Pabitora Wildlife Sanctuary, Assam, India Anwaruddin Choudhury 89 89 History The royal hunt of tiger and rhinoceros in the Nepalese terai in 1911 Kees Rookmaaker, Barbara Nelson and Darrell Dorrington 98 Some observations on the presence of one-horned rhinos in the bas reliefs of the Angkor Wat temple complex in Cambodia Hans H. de Iongh, H.H.T. Prins, N. van Strien and L.G. Rookmaaker 101 Opinion 101 Mineral prospecting in the Selous Game Reserve and its dangers to rhino conservation Rolf D. Baldus 106 Field note 106 Less room for a small population of elephants in severely encroached Mikeno massif, southern Virunga National Park, Democratic Republic of Congo Leonard Mubalama and Déo Mbula 110 Guidelines to contributors Views expressed in Pachyderm are those of the individual authors and do not necessarily reflect those of IUCN, the Species Survival Commission or any of the three Specialist Groups responsible for producing Pachyderm (the African Elephant Specialist Group, the African Rhino Specialist Group and the Asian Rhino Specialist Group). African Elephant Specialist Group report CHAIR REPORTS RAPPORTS DES PRESIDENTS African Elephant Specialist Group report Rapport du Groupe Spécialiste des Eléphants d’Afrique Holly T. Dublin, Chair/Président PO Box 68200, 00200 Nairobi, Kenya; email: [email protected] The latest IUCN quadrennium officially came to a close at the Third IUCN World Conservation Congress held in Bangkok in November 2004. There the membership of all six IUCN Commissions including the Chairs and members of the Species Survival Commission (SSC) Specialist Groups was dissolved and new Chairs were elected to head the IUCN commissions. I am pleased to report that I was elected Chair of the SSC. While my new duties obviously demand an enormous amount of time, I have nevertheless decided to stay on as Chair of AfESG when the appointments take place, and I intend to do so through the current quadrennium. There are obviously some new constraints to the level of day-to-day interaction and supervision that I can provide to steer the group’s work, but thanks to the efforts and talents of the AfESG Secretariat staff, I am glad to note that the core business of AfESG has continued unhindered and at the same frenzied pace as ever. African Elephant Database This issue of Pachyderm includes a paper by AfESG’s Data Review Working Group (DRWG) (p. 19) on the first major analysis of changes in savanna elephant populations in southern and eastern Africa in the period between the African Elephant Database 1998 (Barnes et al. 1999) and the African Elephant Status Report 2002 (Blanc et al. 2003, the AESR. It took the DRWG considerable time and effort to develop a suitable method of analysis. The analysis covers populations of only the two subregions where sur- Pachyderm No. 38 January–June 2005 Les quatre dernières années de l’UICN se sont officiellement clôturées lors du Troisième Congrès Mondial de la Conservation de l’UICN qui s’est tenu à Bangkok en novembre 2004. Les six Commissions de l’UICN furent dissoutes, y compris les Présidents et membres des Groupes Spécialistes de la Commission de Sauvegarde des Espèces (CSS), et de nouveaux Présidents furent élus pour les diriger. Je suis heureuse de vous annoncer que j’ai été élue Présidente de la CSS. Bien que mes nouvelles responsabilités requièrent évidemment un temps énorme, j’ai décidé de rester Présidente du GSEAf lorsque survinrent les nominations, et j’ai l’intention de poursuivre jusqu’à la fin des quatre années en cours. Bien sûr, certaines contraintes nouvelles pèsent sur le niveau des interactions et de la supervision journalières que je pourrai fournir pour diriger le travail du groupe, mais grâce aux efforts et aux talents du staff du Secrétariat du GSEAf, je suis heureuse de pouvoir dire que le travail de base du GSEAf s’est poursuivi sans encombre, au même niveau frénétique que d’habitude. La Base de Données sur l’Eléphant africain Ce numéro de Pachyderm contient un article du Groupe de Travail chargé de la Révision des Données (GTRD) du GSEAf (p. 79) sur la première analyse majeure des changements dans les populations d’éléphants de savane en Afrique australe et de 1 Dublin veys were repeated using similar methods; it excludes considerable portions of range in both subregions as well as all elephant range in West and Central Africa. However, the number of elephants covered in the study does represent a high proportion of the continent’s elephants classified as definite and probable. A meeting of the DRWG was held in Nairobi in November 2004. As well as welcoming newly appointed member John Hart, who will be overseeing the data updates and inputs from Central Africa, the DRWG conducted a thorough review of the process that led to the publication of the AESR 2002 and assessed the feedback received from readers, which has so far been overwhelmingly positive. This review led to agreement on many further improvements for the next edition of the AESR, and the AED manager, Julian Blanc, is currently working to implement them. Data are now being collected for the AESR 2006. In this update cycle, and due to cost considerations, we are trying out an electronic version of the data collection questionnaire. Many of you will have already been contacted by the AED manager and hopefully you will have sent the relevant updates. If you have not done so yet, please point your browser to http://iucn.org/afesg/aed/aedquest/, download the questionnaire file for your country and, once you have completed it, send it to the AED manager at [email protected], together with any survey reports, by 31 December 2005. It is only through the contributions of AfESG members and other collaborators, including the readership of Pachyderm at large, that we can continue to make the AED the most comprehensive and up-to-date single-species database in the world. Your contributions will be duly acknowledged and much appreciated, and they will be integrated into the AESR 2006, which will, subject to funding, appear in print and on the AfESG website sometime in 2006. Updates on subregional and national elephant conservation and management strategies Central Africa Thanks to funding from the World Wide Fund for Nature’s (WWF’s) African Elephant Programme and the Wildlife Conservation Society, in addition to a preexisting and generous contribution made by the Netherlands Committee of IUCN, we are finally in position 2 l’orientale durant la période comprise entre la Base de Données sur l’Eléphant africain 1998 (Barnes et al. 1999) et le Rapport 2002 sur le Statut de l’Eléphant africain (Blanc et al. 2003, le RSEA). Le GTRD a consacré un temps et des efforts considérables à la mise au point d’une méthode d’analyse adéquate. L’analyse ne concerne que les populations des deux sous-régions où les reconnaissances furent répétées en utilisant des méthodes similaires ; elle exclut des portions considérables de l’aire de répartition dans les deux sous-régions de même que toute l’aire de répartition d’Afrique centrale et de l’Ouest. Cependant, le nombre d’éléphants couverts par cette étude représente une proportion élevée des éléphants du continent classés comme certains et probables. Une réunion du GTRD s’est tenue à Nairobi en novembre 2004. En plus de l’accueil de notre nouveau membre, John Hart, qui va superviser la mise à jour des données et les contributions provenant d’Afrique centrale, le GTRD a conduit une révision complète du processus ayant mené à la publication du RSEA 2002 et a évalué le feedback obtenu des lecteurs, qui est jusqu’ici tout à fait positif. Cette révision a mené à un accord sur de nombreuses améliorations pour la prochaine édition du RSEA et Julian Blanc, le gestionnaire de la BDEA, travaille actuellement à les mettre en œuvre. Les données sont actuellement récoltées pour le RSEA 2006. Pour ce cycle de mise à jour, et en raison des coûts, nous testons pour le moment une version électronique du questionnaire de collecte des données. Beaucoup parmi vous auront été déjà contactés par le gestionnaire de la BDEA et j’espère que vous lui avez déjà envoyé les mises à jour en question. Si vous ne l’avez pas encore fait, pointez votre navigateur web sur http://iucn.org/afesg/aed/aedquest/, chargez le document de questionnaire pour votre pays et, lorsque vous l’aurez complété, envoyez-le au gestionnaire de la BDEA à [email protected] avec tout autre rapport de reconnaissance, pour le 31 décembre 2005. Ce n’est que grâce aux contributions des membres du GSEAf et d’autres collaborateurs, y compris l’ensemble des lecteurs de Pachyderm, que nous pouvons continuer à faire de la BDEA la base de données la plus complète et la plus actualisée du monde pour une seule espèce. Vos contributions seront très appréciées et dûment reconnues, et elles seront intégrées dans le RSEA 2006 qui sera publié, si les finances le permettent, en version papier et sur le site du GSEAf, dans le courant de 2006. Pachyderm No. 38 January–June 2005 African Elephant Specialist Group report to move forward with developing the Central Africa Elephant Conservation Strategy (CAECS). The AfESG Secretariat is now busy working on the administrative and logistical preparations for a strategic planning workshop scheduled to take place in early September 2005 in Cameroon. A comprehensive background document on the status of elephants in Central Africa and the main threats and opportunities for their conservation is also being compiled. IUCN’s Central Africa Programme Office will facilitate the workshop, which will be attended by wildlife directors of all seven African elephant range states of Central Africa as well as key NGOs, donors and private sector partners. Illegal killing of elephants for ivory is one of the issues expected to feature as a priority in the subregional strategy. In particular, the role played by unregulated domestic ivory markets has been singled out as an issue requiring immediate attention. The range states’ resolve to shut down these markets is currently being tested by the CITES Action Plan for the control of trade in African elephant ivory (CITES Decision 13.26), which outlines a number of activities that all African elephant range states are required to carry out to curb domestic ivory markets. Any country that fails to comply with the action plan could face sanctions on all commercial trade in specimens of CITES-listed species. Several countries in Central Africa are directly implicated in the CITES decision. Another major threat to elephants in Central Africa is the elephant meat trade, which is often exacerbated by infrastructure development, the growing presence of extractive and exploitative industries and easy access to weapons. In the absence of effective law enforcement, elephants are often the first large mammals to fall prey to poachers, as they yield a higher amount of meat per bullet than any other species and provide the added bonus of ivory as a lucrative by-product. The CAECS discussions will provide a useful platform from which to start building partnerships with private sector partners, such as the logging industry, whose activities can have a large indirect effect on the elephant meat trade. In addition the CAECS should strengthen its synergy with other relevant subregional processes such as the Conference of Ministers in Charge of Forests in Central Africa (COMIFAC) and the Africa Forest Law Enforcement and Governance forum. On a national scale, AfESG was recently approached to advise on a strategic planning exercise Pachyderm No. 38 January–June 2005 Nouvelles sur les stratégies sousrégionales et nationales de conservation et de gestion des éléphants Afrique centrale Grâce au financement du Programme des Eléphants d’Afrique du Fonds Mondial pour la Nature (WWF) et de la Wildlife Conservation Society, venant s’ajouter à une contribution généreuse du Comité néerlandais de l’UICN, nous sommes finalement prêts à lancer le développement de la Stratégie de Conservation des Eléphants d’Afrique centrale. Le Secrétariat du GSEAf s’occupe de la préparation administrative et logistique d’un atelier de planification stratégique qui devrait avoir lieu début septembre 2005 au Cameroun. Un document d’information complet sur le statut des éléphants en Afrique centrale, sur les menaces principales et les opportunités pour leur conservation est aussi en préparation. Le Bureau Régional de l’UICN pour l’Afrique centrale va faciliter l’atelier auquel participeront les directeurs de la faune des sept états de l’aire de répartition des éléphants en Afrique centrale, ainsi que les ONG, les bailleurs de fonds et les partenaires du secteur privé. L’abattage illégal des éléphants pour l’ivoire est un des sujets qui devrait être prioritaire dans la stratégie sous-régionale. Le rôle joué par les marchés domestiques non réglementés a été particulièrement pointé du doigt comme exigeant une attention immédiate. La détermination des états de l’aire de répartition à fermer ces marchés est actuellement testée par le Plan d’Action de la CITES pour le contrôle du commerce d’ivoire de l’éléphant africain (Décision CITES 13.26), qui décrit un certain nombre d’activités que tous les Etats de l’aire de répartition des éléphants d’Afrique sont tenus d’entreprendre pour juguler les marchés domestiques d’ivoire. Tout pays qui n’applique pas ce plan d’action pourrait se voir infliger des sanctions sur tout le commerce de spécimens d’espèces listées dans la CITES. Quelques pays d’Afrique centrale sont directement concernés par cette décision de la CITES. Une autre menace majeure sur les éléphants d’Afrique centrale est le commerce de viande d’éléphant qui est souvent accru par le développement d’infrastructures, par la présence croissante d’industries extractives et d’exploitation et par la disponibilité aisée d’armes. En absence d’une application effective des 3 Dublin for elephant conservation in the Democratic Republic of Congo, where the internal political situation and continuing insecurity are posing major challenges for conservation of the country’s remaining elephants. We are now awaiting further guidance from our colleagues in DRCongo on exactly when and how they would like us to engage in this important initiative. Southern Africa At the request of the range state governments of southern Africa, AfESG continued to advise as a regional elephant conservation strategy is being developed. A three-day strategic planning workshop is to be held in the near future to start fleshing out the strategy. At this workshop AfESG will be presenting an overview on the status of elephants in the region and help provide technical input to the strategic planning. Much of the discussion at the upcoming workshop is expected to focus on the growing challenges of local overpopulation of elephants in southern Africa. While a number of options are available for dealing with this problem, ultimately the decision on which is chosen will likely be complex, involving both subjective and objective considerations. AfESG’s job will be to help ensure that the range states are given the technical advice they need to enable them to evaluate the pros and cons of the different options and to provide technical advice when the selected strategies are subsequently implemented. West Africa It is expected that the West African Elephant Conservation Strategy, which forms the central operational component of a draft intergovernmental memorandum of understanding (MOU) between West African states on conserving elephants in the region, will be endorsed by the relevant ministers from the West African range states at the Conference of the Parties to the Convention on Migratory Species in November 2005 in Nairobi. This high-level recognition and endorsement is gratifying after all the years of investment. It is our hope and belief that the signing of this MOU will open a new chapter for conserving elephants in West Africa. The largest and most important elephant populations in West Africa use areas that straddle the boundaries of two or more countries. This presents special challenges for their conservation and requires 4 lois, les éléphants sont souvent les premiers grands mammifères à être la proie des braconniers car ils procurent une plus grande quantité de viande par munition que n’importe quelle autre espèce et fournissent en prime de l’ivoire comme sous-produit lucratif. Les discussions sur la Stratégie de Conservation des Eléphants d’Afrique Centrale (SCEAC) fourniront une plateforme utile en vue de bâtir des partenariats avec le secteur privé, telle l’industrie forestière dont les activités peuvent avoir un effet indirect considérable sur le commerce de viande d’éléphant. De plus, la SCEAC devrait renforcer sa synergie avec d’autres processus sousrégionaux pertinents tels que la Conférence des Ministres en charge des Forêts d’Afrique Centrale (COMIFAC) et le forum sur l’Application de la Législation Forestière et Gouvernance en Afrique. A l’échelle nationale, le GSEAf fut récemment approché pour fournir des conseils sur un exercice de planification stratégique pour la conservation des éléphants en République Démocratique du Congo où la situation politique intérieure et l’insécurité persistante posent des défis majeurs à la conservation des éléphants subsistant dans le pays. Nous attendons à présent des instructions de nos collègues en RDC pour savoir exactement quand et comment ils souhaiteraient que nous nous engagions dans cette importante initiative. Afrique australe A la demande des gouvernements des états de l’aire de répartition, le GSEAf a continué de fournir des conseils sur le développement d’une stratégie sousrégionale de conservation des éléphants pour l’Afrique australe. Un atelier de planification stratégique de trois jours doit se tenir très bientôt pour élaborer cette stratégie. Lors cet atelier, le GSEAf va présenter une revue du statut des éléphants dans la sous-région et contribuera à fournir un appui technique à la planification stratégique. Lors de cet atelier, on s’attend à ce qu’une grande partie de la discussion se concentre sur les défis croissants liés à la surpopulation locale des éléphants en Afrique australe. S’il existe toute une variété d’options pour faire face à ce problème, celle qui sera finalement choisie sera probablement complexe et impliquera des considérations tant subjectives qu’objectives. Le tâche pour le GSEAf sera de s’assurer que les états de l’aire de répartition disposent des conseils techniques dont ils ont besoin pour Pachyderm No. 38 January–June 2005 African Elephant Specialist Group report close cross-border collaboration between neighbouring range states. In an effort to encourage such collaboration, AfESG convened a workshop in June 2003 to develop action plans for five of the main crossborder elephant conservation areas in West Africa. Subsequently, AfESG has assisted in coordinating and implementing many of the activities in these action plans. Most recently, funds were successfully raised to carry out activities on both Burkina Faso and Ghanaian sides of the Kabore Tambi–Red Volta–Doung elephant corridor. A proposal was also written to develop an action plan for the Ziama–Northeast Forest Reserve area, which straddles the borders of Guinea Conakry and Liberia. The Keidaren Nature Conservation Fund of Japan has recently agreed to contribute funds towards this initiative—the first contribution from a Japanese NGO to AfESG! With Benin as the latest country to have embarked on developing a national strategy for elephant conservation, 10 of the 13 West African range states now have national elephant conservation strategies—a significant feat considering that until a few years ago no such strategy existed, and West African elephants generally received little local or international attention. The Benin strategic planning workshop, which took place in early April 2005, was attended by all major stakeholders including Lamine Sebogo, the AfESG programme officer for West Africa, who was involved in this process from its inception. Eastern Africa The Kenya Wildlife Service has secured some funding from WWF’s Africa Elephant Programme for an extensive, proposed participatory process to develop a national elephant conservation strategy, and discussions are currently under way with other interested donors. AfESG, which helped KWS draft the funding proposal, is ready to provide further technical advice in developing the strategy and subsequently implementing it if requested to do so. Human–elephant conflict Developing national systems for managing human–elephant conflict AfESG has recently embarked on an ambitious project to help develop national models for managing human–elephant conflict (HEC). To be effective, such Pachyderm No. 38 January–June 2005 évaluer les avantages et inconvénients des différentes options, et de fournir des avis techniques lorsque les stratégies sélectionnées seront mises en œuvre. Afrique de l’Ouest On s’attend à ce que la Stratégie de Conservation des Eléphants d’Afrique de l’Ouest, qui constitue la composante opérationnelle centrale d’un projet d’Accord intergouvernemental entre les états d’Afrique de l’Ouest sur la conservation des éléphants dans la sous-région, soit entérinée par les ministres des états de l’aire de répartition d’Afrique de l’Ouest, à la Conférence des Parties à la Convention sur les Espèces Migratrices, en novembre 2005, à Nairobi. Cette reconnaissance à haut niveau et cette adhésion sont très satisfaisantes après toutes ces années d’investissement. Nous espérons et nous croyons que la signature de cet Accord ouvrira une nouvelle ère pour la conservation des éléphants en Afrique de l’Ouest. Les plus grandes et les plus importantes populations d’éléphants d’Afrique de l’Ouest parcourent des zones qui s’étendent de part et d’autres de frontières internationales. Ceci constitue un défi spécial pour leur conservation et exige une collaboration transfrontalière étroite entre états voisins. Afin d’encourager une telle collaboration, le GSEAf a organisé un atelier en juin 2003 pour développer un plan d’action pour cinq des principales zones transfrontalières de conservation des éléphants en Afrique de l’Ouest. Après cela, le GSEAf a contribué à la coordination et à la mise en œuvre de beaucoup d’activités inscrites dans ces plans d’action. Tout récemment, des fonds furent récoltés pour entreprendre des activités dans le corridor Kabore Tambi–Volta Rouge–Doung, tant du côté burkinabé que du côté ghanéen. Nous avons aussi rédigé une proposition visant à développer un plan d’action pour la région de Ziama et la Réserve Forestière de Nordest, qui s’étend des deux côtés de la frontière entre la Guinée Conakry et le Liberia. Le Keidaren Nature Conservation Fund japonais a récemment accepté de contribuer au financement de cette initiative–la première contribution d’une ONG japonaise au GSEAf ! Avec le Bénin qui est le dernier pays à s’être attelé au développement d’une stratégie nationale pour la conservation des éléphants, 10 des 13 états de l’aire de répartition ouest-africains ont à présent une stratégie nationale de conservation des éléphants—une réussite 5 Dublin systems will need to take a holistic approach that involves a diverse set of actors at all levels, from the affected community up to the relevant local, district and national government policy-makers. AfESG will be developing and testing appropriate actions at each of these levels. This will of course require a significant amount of time and funding, and we are currently working on a detailed financing proposal for a five-year medium-sized project under the Global Environment Facility of the United Nations Development Programme (UNDP GEF). If approved, it would provide up to USD 1 million over five years, or 50% of the proposed budget of USD 2 million. The idea of piloting this approach received broad support from the range states at the recent African Elephant Range States Dialogue meeting that took place in Bangkok, Thailand, in September 2004. AfESG was invited to attend a workshop on human–wildlife conflict management (HWCM) in Namibia, which took place 16–17 May 2005, and was funded by the United States Agency for International Development (USAID) as part of the UNDP GEFsupported project Strengthening the Protected Areas Network (SPAN). The objectives of this workshop are as follows: to develop a framework for future HWCM policy directions in Namibia; to initiate the development of a standardized monitoring system for HWCM; to discuss best-practice mitigation measures in Namibia and throughout the region; and to launch a survey on HWC situations in two areas. AfESG has been requested to give presentations on the key lessons learned from our study of the nature and mitigation of human–elephant conflict across the continent and to help facilitate some of the working group sessions. Curriculum and training modules for HEC management We are in the final stages of negotiating funding for a project to develop AfESG-certified curriculum and training modules for HEC management. This project will culminate in a training workshop, sometime in 2006, in which the new training modules will be used to train HEC practitioners from different conflict sites in at least two countries. The curriculum is expected to give the trainees all the background and tools needed to develop effective HEC-mitigation strategies, as well as to help them train others to effectively manage a variety of different conflict situations. The project will 6 remarquable si l’on pense qu’aucune stratégie de ce genre n’existait seulement il y a quelques années, et que les éléphants d’Afrique de l’Ouest ne recevaient généralement que peu d’attention au niveau local ou international. L’atelier de planification stratégique du Bénin qui s’est déroulé en avril 2005, a vu la participation de tous les acteurs majeurs, y compris Lamine Sebogo, le coordinateur du programme du GSEAf pour l’Afrique de l’Ouest, qui est impliqué dans ce processus depuis son démarrage. L’Afrique orientale Le Kenya Wildlife Service (KWS) a obtenu quelques fonds du Programme des Eléphants d’Afrique du WWF pour un vaste projet de processus participatif pour développer une stratégie nationale de conservation des éléphants, et des discussions sont en cours avec d’autres donateurs. S’il y est invité, le GSEAf, qui a aidé KWS à préparer la proposition de financement, est prêt à fournir encore un appui technique pour développer, et ensuite pour mettre en œuvre la stratégie. Conflit homme–éléphant Le développement de systèmes nationaux pour gérer les conflits homme–éléphant Le GSEAf s’est récemment embarqué dans un projet ambitieux visant à développer des modèles nationaux pour gérer les conflits homme–éléphant (CHE). Pour être efficaces, de tels systèmes devront adopter une approche holistique qui implique une palette diversifiée d’acteurs à tous les niveaux, depuis la communauté affectée jusqu’aux preneurs de décisions aux niveaux locaux, régionaux et nationaux pertinents. Le GSEAf va développer et tester les actions appropriées à chacun de ces niveaux. Ceci va bien sûr requérir une quantité considérable de temps et de fonds, et nous sommes en train de travailler à une proposition de financement détaillée pour un projet de taille moyenne de cinq ans à soumettre au Fonds pour l’Environnement Mondial du Programme des Nations unies pour le Développement (PNUD-FEM). S’il était approuvé, il fournirait jusqu’à un million de USD pour cinq ans, soit 50 % du budget proposé qui est de 2 millions de USD. L’idée de piloter une telle approche a reÁu un support massif des états de l’aire de répartition au cours de la réunion de Dialogue entre Etats de l’aire de répartition des éléphants africains qui eut lieu à Bangkok, en Thaïlande, en septembre 2004. Pachyderm No. 38 January–June 2005 African Elephant Specialist Group report be implemented in close collaboration with the Elephant Pepper Development Trust, headed by AfESG member Dr Loki Osborn, whose facility in Zambia will also be used to hold the HEC training course. Forging closer links on human–elephant conflict work with IUCN in Africa AfESG has begun discussions with IUCN regional and national offices in Africa about increasing the synergy between AfESG’s activities and IUCN’s programmes. This is in line with new efforts to ensure effective delivery by the commissions and the thematic and regional programmes towards the institutional goals and objectives of IUCN. One realm of mutual interest is the need to manage HEC while simultaneously conserving biodiversity and improving local livelihoods. Some IUCN projects where AfESG’s technical expertise could be brought to bear to deal with HEC problems have already been identified, and discussions are under way with relevant IUCN staff about the practical aspects of AfESG involvement. Similar synergy may be possible with other partner organizations. Update on the CITES MIKE programme Like AfESG, the Central Coordinating Unit of the CITES MIKE programme has recently been focusing all its energies on trying to find sufficient funding to be able to maintain its core operations. There is now hope that a significant amount of funding may be made available from the resources of the European Commission-administered European Development Fund (EDF) for the Africa, Caribbean and Pacific (ACP) regions. In April MIKE’s financing proposal for a five-year ACP/EDF project was assessed by an independent team, and the proposal is now ready for final submission to the European Union. If this proposal is approved, hopefully the funds will become available in early 2006. Until this longterm funding is secured, MIKE will have to find bridging funds to remain operational. Collaborative activities in the proposal that would directly involve AfESG include a study to determine the impact of the elephant meat trade on elephant populations in Central Africa, and closer MIKE-AED Pachyderm No. 38 January–June 2005 Le GSEAf a été invité à participer à un atelier sur la gestion des conflits homme–faune sauvage (CHFS) en Namibie les 16–17 mai 2005, qui est financé par l’Agence des Etats-Unis pour le Développement International (USAID), en tant que partie du projet PNUD-FEM pour le Renforcement du Réseau des Aires Protégées (SPAN). Les objectifs de cet atelier sont les suivants : développer un cadre pour les futures directives politiques de CHFS en Namibie ; initier le développement d’un système de monitoring standardisé pour les CHFS ; discuter les meilleures mesures de mitigation en Namibie et dans toute la région ; et lancer une reconnaissance des situations HFS dans eux zones. On a demandé au GSEAf de présenter les leçons clés tirées de notre étude sur la nature et la mitigation des conflits homme–éléphant dans tout le continent et d’aider à faciliter certaines des sessions des groupes de travail. Curriculum et modules de formation pour la gestion des CHE Nous sommes au stade final de la négociation d’un financement pour un projet visant à développer un curriculum certifié par le GSEAf sur la gestion des CHE. Ce projet va se terminer au cours de 2006 par un atelier de formation durant lequel les nouveaux modules de formation seront utilisés pour former les praticiens des CHE provenant de différents sites à conflits dans au moins deux pays. Le curriculum devrait fournir aux stagiaires tout le contexte et les outils nécessaires pour développer des stratégies efficaces de mitigation des CHE, et aussi les aider à en former d’autres à la gestion de toute une variété de situations de conflit différentes. Le projet sera mis en œuvre en collaboration étroite avec Elephant Pepper Development Trust, qui est dirigé par un membre du GSEAf, le Dr Loki Osborn, dont les installations en Zambie serviront également à réaliser la formation en CHE. Tisser des liens plus étroits avec l’UICN en Afrique pour le travail sur les conflits homme–éléphant Le GSEAf a entamé des discussions avec les bureaux régionaux et nationaux de l’UICN en Afrique au sujet du renforcement de la synergie entre les activités du GSEAf et les programmes de l’UICN. Ceci répond aux efforts nouveaux destinés à assurer une contribution efficace des commissions et des programmes 7 Dublin integration. In fact, seven elephant population censuses—six in Central Africa (Bangassou, Boumba Bek, Dzanga-Sangha, Minkébé, Nouabalé-Ndoki, Salonga) and one in Guinea Conakry in West Africa— have just been completed under the aegis of the MIKE programme. These surveys will contribute towards fulfilling the CITES decision to establish a baseline against which to determine trends in illegal killing and to analyse the factors involved. These surveys will also provide much-needed information, some of it unprecedented, on the status of key elephant populations in these subregions. thématiques et régionaux aux buts et objectifs institutionnels de l’UICN. Un domaine d’intérêt mutuel est la nécessité de gérer les conflits homme–éléphant tout en conservant simultanément la biodiversité et en améliorant les moyens d’existence locaux. On a déjà identifié quelques projets de l’UICN où l’expertise technique du GSEAf pourrait être utilisée pour s’occuper de problèmes liés aux CHE, et des discussions avec les responsables concernés de l’UICN sont en cours sur les aspects pratiques de l’implication du GSEAf. Une synergie similaire devrait être possible avec d’autres organisations partenaires. The Local Overpopulation Task Force Nouvelles du programme CITES MIKE Thanks to support from WWF-Switzerland and the Toronto Zoo, AfESG now has enough funds to convene a workshop to help develop the technical guidelines on best practice for managing local overpopulation problems. AfESG’s Local Overpopulation Task Force is already working on a first draft, which will be edited by the AfESG Secretariat and will form the basis of the discussion at the workshop. Dates have not yet been finalized, but it is hoped that the workshop will take place before the end of the year. Tout comme le GSEAf, l’Unité de Coordination Centrale du programme CITES MIKE a concentré récemment toute son énergie à la recherche d’un financement suffisant pour permettre le maintien de ses opérations de base. L’espoir existe à présent qu’une part significative du financement soit rendue disponible grâce aux ressources du Fonds Européen de Développement (FED) pour les régions Afrique– Caraïbes–Pacifique administré par la Commission européenne. En avril, la proposition de financement de MIKE pour un projet FED/ACP de cinq ans fut évaluée par un team indépendant, et la proposition est à présent prête pour sa soumission finale à l’Union Européenne. Si cette proposition est approuvée, les fonds devraient être disponibles début 2006. En attendant que ce financement à long terme soit sécurisé, MIKE devra continuer à trouver des financements de raccord afin de rester opérationnel. Parmi les activités collaboratives de la proposition qui impliqueraient directement le GSEAf, il y a une étude pour déterminer l’impact du commerce de viande d’éléphant sur les populations d’éléphants en Afrique centrale, ainsi qu’une intégration plus étroite entre MIKE et la BDEA. En fait, sept recensements de populations d’éléphants—six en Afrique centrale (Bangassou, Boumba Bek, Dzanga-Sangha, Minkébé, Nouabalé-Ndoki, Salonga) et un en Guinée Conakry, en Afrique de l’Ouest—viennent d’être achevés sous l’égide du programme MIKE. Ces reconnaissances devraient répondre à la décision de la CITES d’établir la base par rapport à laquelle seront déterminées les tendances de l’abattage illicite et l’analyse des facteurs impliqués. Ces reconnaissances fourniront également The AfESG website The AfESG website http://iucn.org/afesg has been given a totally new look, loosely based on the current IUCN and SSC websites. Other changes made include adding links on every English page to the same page in French, and vice versa, and including a ‘you are here’ navigation bar along the top of each page. Cascading style sheets, which ensure a consistent format throughout the site and make it easier to create consistent-looking web pages, were also developed. The French version of the African Elephant Status Report 2002 and issue 37 of Pachyderm were added to the ever-growing list of resources on this popular website. That our continuing efforts to develop our website into an effective tool to disseminate information on African elephant conservation and management has paid off is demonstrated by the fact that the site now consistently receives over 2000 hits per day, almost double the rate a year ago! 8 Pachyderm No. 38 January–June 2005 African Elephant Specialist Group report The African Elephant Library During this period several additional documents were collected for the African Elephant Library and stored either electronically or in hard copy. Thanks to support from Save the Elephants, a new system has been put in place to facilitate cataloguing and annotating. The library now has over 4700 references and is growing by the day. Prospects for the future It had been my sincerest hope that this time round I would be able to report some good news about the long-term funding prospects for AfESG. Unfortunately, despite trying every avenue known to us, we have still not been able to source the funds needed to give us the stability and security to implement many of the priority actions needed to fulfil AfESG mission and objectives. It is only thanks to the continuing support of the US Fish and Wildlife Service and the UK Department for Food, Environment and Rural Affairs to our core operating costs that AfESG has been able to function at all these past six months, and we are hoping for further support from these same two donors to see us through to the end of 2005. The only other ray of hope at present comes in the form of a possible contribution by the French government to help cover some of the core operating costs of the AfESG programme office for West Africa. As this update demonstrates, we have continued to be very productive despite the ongoing uncertainties about future funding. The short-term nature of our current bridging grants means that the pressure on the AfESG Secretariat to continue fund raising is likely to remain high into the foreseeable future. So, stay tuned for our next update and wish us good luck in our efforts! References Barnes RFW, Craig CG, Dublin HT, Overton G, Simon W, Thouless CR. 1999. African Elephant Database 1998. IUCN/SSC African Elephant Specialist Group, Gland, Switzerland. Blanc JJ, Thouless CR, Hart JA, Dublin HT, Douglas-Hamilton I, Craig CG, Barnes RFW. 2003. African elephant status report 2002: an update from the African Elephant Database. IUCN/SSC African Elephant Specialist Group, Gland, Switzerland. Pachyderm No. 38 January–June 2005 une information indispensable, et dans certains cas sans précédent, sur le statut des populations clés d’éléphants dans ces sous-régions. Groupe Spécial sur la surpopulation locale Grâce au support du WWF Suisse et du Zoo de Toronto, le GSEAf a maintenant assez de fonds pour convoquer un atelier visant à soutenir le développement de lignes directrices sur les meilleures pratiques pour gérer les problèmes de surpopulation locale. Le Groupe Spécial sur la Surpopulation locale du GSEAf travaille déjà sur un premier texte qui sera publié par le Secrétariat du GSEAf et formera la base de discussion lors de l’atelier. Les dates n’ont pas encore été finalisées, mais on espère que l’atelier aura lieu avant la fin de l’année. Le site web du GSEAf Nous avons donné au site web du GSEAf http:// iucn.org/afesg un look totalement neuf, vaguement inspiré des sites web actuels de l’UICN et de la CSS. D’autres changements réalisés comprennent l’addition de liens entre chaque page en anglais vers la même page en français, et vice versa, ainsi que l’inclusion en tête de chaque page d’une barre de navigation ‘vous êtes ici’. On a aussi développé des pages de même style en cascade qui assurent un format cohérent à travers tout le site et permettent de garder une cohésion générale au site. La version française du Rapport 2002 sur le Statut de l’Eléphant africain 2002 et le numéro 37 de Pachyderm ont été ajoutés à la liste toujours plus longue des ressources de ce site web populaire. Nos efforts continuels pour faire du site Internet un outil efficace pour disséminer l’information sur la conservation et la gestion de l’éléphant africain se sont révélés payants; on en veut pour preuve le fait que ce site reçoit régulièrement plus de 2000 visites par jour, soit près du double de l’an passé ! La bibliothèque sur l’éléphant africain Nous avons collecté au cours de cette période plusieurs documents supplémentaires que nous avons classés en version électronique ou en version papier dans la Bibliothèque sur l’éléphant africain. Grâce à 9 Dublin l’appui de Save the Elephants, on a mis en place un nouveau système pour les cataloguer et les annoter plus facilement. La Bibliothèque compte aujourd’hui plus de 4700 références, et la liste s’allonge chaque jour. Perspectives d’avenir J’avais réellement espéré pouvoir apporter cette fois quelque bonne nouvelle à propos des perspectives d’avenir pour le financement du GSEAf. Malheureusement, après avoir exploré toutes les voies possibles à notre connaissance, nous ne sommes toujours pas en mesure d’identifier les financements nécessaires pour nous assurer la stabilité et la sécurité nécessaires pour mettre en œuvre les nombreuses activités prioritaires requises pour remplir la mission et les objectifs du GSEAf. Ce n’est que grâce au support continu du US Fish and Wildlife Service et du Department for Food, Environment and Rural Affairs de Grande Bretagne pour nos coûts opérationnels de base, que le GSEAf a été en mesure de fonctionner au cours des six mois passés. Nous espérons un appui supplémentaire de ces deux bailleurs pour arriver fin d’année 2005. La seule autre lueur d’espoir actuelle est la perspective d’une contribution possible du gouvernement français pour contribuer à certains coûts opérationnels du bureau de programme du GSEAf pour l’Afrique de l’Ouest. Comme le montre cette mise à jour, nous avons continué à être très productifs en dépit des incertitudes présentes sur l’avenir de notre financement. Le caractère de court terme de nos actuelles subventions de raccord signifie que la pression qui pèse sur le Secrétariat du GSEAf pour poursuivre la recherche de fonds va probablement rester importante dans l’avenir immédiat. Dès lors, restez à l’écoute de notre prochaine mise à jour et souhaitez-nous bonne chance pour nos efforts ! Références Barnes RFW, Craig GC, Dublin HT, Overton G, Simons W, Thouless CR. 1999. African Elephant database 1998. IUCN/SSC African Elephant Specialist Group, Gland, Switzerland. Blanc JJ, Thouless CR, Hart JA, Dublin HT, Douglas-Hamilton I, Craig CG, Barnes RFW. 2003. African elephant status report 2002: an update from the African Elephant Database. IUCN/SSC African Elephant Specialist Group, Gland, Switzerland. 10 Pachyderm No. 38 January–June 2005 African Elephant Specialist Group report African Rhino Specialist Group report Rapport du Groupe Spécialiste des Rhinos d’Afrique Martin Brooks, Chair/Président 59 Silverdale Crescent, Chase Valley, Pietermaritzburg 3201, South Africa email: [email protected] Rhinos remain Critically Endangered Les rhinos restent en danger critique d’extinction In my last Chair report I concentrated on reporting on the seventh AfRSG meeting. Sadly, the status of the Critically Endangered northern white rhino Ceratotherium simum cottoni has continued to decline, and there are now probably fewer than 10 individuals of this subspecies in the wild, in Garamba National Park in the Democratic Republic of Congo (DRC). Its future hangs in the balance as the survival strategy agreed with the DRC government has stalled due to internal problems. The escalation of commercial poaching that precipitated the current crisis was the catalyst for a stakeholders’ workshop, held by the Institut Congolais pour la Conservation de la Nature (ICCN) and partners in July 2004.This workshop recommended 1) an emergency programme of prioritized support for Garamba, and concurrently 2) the capture and relocation of at least five rhinos to a more secure area pending their later return. Subsequent site evaluations identified Ol Pejeta Ranch in Kenya as the preferred site. Then in January 2005, I received an invitation from the Administrateur Délégué Général of ICCN to visit Kinshasa to lead a delegation of stakeholders for meetings in mid-January to discuss the strategy and seek approval from the highest level within government. Apart from AfRSG, the delegation comprised the International Rhino Foundation, Fauna and Flora International, UNESCO World Heritage, IUCN Central Africa, the World Bank and the Garamba Project. Meetings were held with, inter alia, the Minister of Environment and Vice Presidents Zaidi Ngoma and Abdulaye Iherodia, and the latter advised us that the Presidential Office supported and had approved the strategy. Very unfortunately the situation deteriorated after our departure, and despite several internal initiatives, the political climate, at least up to mid-April, had not allowed either the formal signing of the agreement or preparations for the translocation to take place. Dans mon dernier Rapport du Président, je me suis concentré sur le rapport de la 7ème Réunion du GSRAf. Malheureusement, le statut du Rhinocéros blanc du Nord (Cerathotherium simum cottoni), « en danger critique d’extinction », a continué à se détériorer et il reste probablement à ce jour moins de 10 individus de cette sous-espèce dans la nature, au Parc National de la Garamba, en République Démocratique du Congo (RDC). Son futur est incertain étant donné que la stratégie de survie qui avait obtenu l’accord du Gouvernement congolais a été stoppée suite à des problèmes internes. L’escalade du braconnage commercial qui a précipité la crise actuelle a été le catalyseur d’un atelier des parties prenantes qui a rassemblé l’Institut congolais pour la Conservation de la Nature (ICCN) et ses partenaires en juillet 2004. Cet atelier a recommandé 1) un programme d’urgence de soutien prioritaire à la Garamba et en même temps 2) la capture et le transfert d’au moins cinq des derniers rhinos dans une zone plus sûre en attendant leur retour. Les évaluations de sites qui ont suivi ces recommandations ont identifié Ol Pejeta Ranch, au Kenya, comme l’endroit préférentiel. Puis, en janvier 2005, j’ai reçu une invitation de l’Administrateur Délégué Général de l’ICCN pour me rendre à Kinshasa et conduire à la mi-janvier 2005 une délégation des parties prenantes pour discuter de la stratégie et solliciter l’accord au plus haut niveau du Gouvernement. En plus du GSRAf, la Délégation comprenait l’International Rhino Foundation, Fauna and Flora International, le Patrimoine Mondial de l’UNESCO, UICN Afrique centrale, la Banque mondiale et le Projet Garamba. Il y eut des réunions avec, entre autres, le Ministre de l’Environnement et les Vice-Présidents Zaidi Ngoma et Abdulaye Yerodia et ces derniers nous ont appris que le Bureau du Président soutenait et avait approuvé la stratégie. Malheureusement, la situ- Pachyderm No. 38 January–June 2005 11 Brooks The other African rhino taxon on the brink of extinction is the West African black rhino, Diceros bicornis longipes. No surveys have yet come up with tangible evidence of the presence of the absolute minimum of five unrelated rhinos (at least three females and one male) required for pursuing a strategy other than simply protecting them in the scattered locations where they currently occur. A fresh initiative is needed that involves all interested parties to determine the potential viability of the remaining population, and I have contacted the Ministry of Environment and Forests in Cameroon in this regard. Moves are currently under way to reintroduce white rhinos to the wild in Uganda. Four southern white rhinos sourced from Kenya are shortly to be introduced to Ziwa Ranch in Uganda. It is also hoped that two other animals currently in Entebbe Zoo will also be translocated to boost founder numbers in Ziwa. During the reporting period a number of discussions were held, particularly through the auspices of the Southern Africa Development Community (SADC) Regional Programme for Rhino Conservation (RPRC), in an attempt to source additional founder rhinos to boost the re-established population of five animals in North Luangwa National Park, Zambia. Encouragingly, it is likely that additional black rhinos will be sourced to boost this population, and I hope to be able to report positively on this matter in the next issue of Pachyderm. It is likely that Botswana’s recently re-established black rhino population will also be boosted in the future. Law enforcement database software Good progress has been made with enhancing the law enforcement database software application ‘Wildlife Investigator’. At a recent SADC Rhino and Elephant Security Group meeting held in the Great Fish River Reserve in Eastern Cape, South Africa, and with SADC RPRC funding, the latest version of the software was distributed to delegates and a hands-on course on its use was given by the AfRSG Scientific Officer assisted by SANPark’s Sandra Snelling and Ken Maggs. Delegates from 17 different agencies (including Interpol’s subregional bureau) from seven countries attended the course. With SADC RPRC funding, Scene of the Crime training has continued with a course being held in Zam- 12 ation s’est détériorée après notre départ et malgré plusieurs initiatives internes, le climat politique, en tout cas jusqu’à mi-avril, n’avait encore permis ni la signature officielle de l’accord ni la possibilité de préparer le transfert des rhinos. L’autre taxon de rhinos africains au bord de l’extinction est le Rhino noir d’Afrique de l’Ouest (Diceros bicornis longipes). Aucune recherche n’a pu apporter la preuve de la présence de cinq rhinos non apparentés (au moins trois femelles et un mâle), nombre qui est le minimum absolu pour poursuivre une stratégie autre que leur simple protection dans les endroits isolés où ils vivent actuellement. Il est nécessaire d’employer une autre stratégie qui implique toutes les parties intéressées afin de déterminer la viabilité potentielle de la population restante, et j’ai contacté à ce sujet le Ministère de l’Environnement et des Forêts au Cameroun. Des démarches sont actuellement en cours pour réintroduire des rhinos blancs dans la nature en Ouganda. Quatre rhinos blancs du Sud provenant du Kenya doivent être relâchés sous peu dans le Ziwa Ranch en Ouganda. On espère que deux autres animaux qui se trouvent aujourd’hui au Zoo d’Entebbe seront aussi transférés pour augmenter le nombre de reproducteurs à Ziwa. Au cours de la période couverte par ce rapport, il y eut de nombreuses discussions, spécialement du fait du Programme Régional pour la Conservation des rhinos (PRCR) de la Southern Africa Development Community (SADC) afin de repérer des rhinos reproducteurs supplémentaires pour renforcer la population de cinq individus que l’on a réintroduite dans le Parc National de Luangwa-Nord, en Zambie. Il est heureusement probable que de nouveaux rhinos noirs seront découverts pour renforcer cette population et j’espère pouvoir vous donner des nouvelles positives à ce sujet dans le prochain numéro de Pachyderm. Logiciel de la base de données sur l’application des lois L’application « Wildlife Investigator » du logiciel de la base de données sur l’application des lois a fait de grands progrès. Lors d’une réunion récente du Groupe chargé de la Sécurité des Rhinos et des Eléphants de la SADC qui s’est tenue dans la Great Fish River Reserve, au Cap oriental, en Afrique du sud, avec un financement du PRCR/SADC, la dernière version du logiciel a été Pachyderm No. 38 January–June 2005 African Rhino Specialist Group report African Elephant bia, and courses were scheduled for both Tanzania and South Africa in May 2005. Kaziranga celebrates 100 years The AfRSG Scientific Officer attended Kaziranga National Park’s centenary celebrations in Assam, India. He gave an invited presentation at a workshop on grassland management in the park, which was held as part of its centenary celebrations. His presentation drew on experience in Africa to promote the desirability for active biological management of rhinos to ensure a rapid growth in numbers, which is beneficial in terms of providing both genetic viability and a greater buffer against poaching. This stimulated subsequent debate in plenary. His visit may prove to be the start of increased sharing of knowledge and ideas between Africa and Asia for the benefit of rhino conservation in the years to come. SADC Regional Programme for Rhino Conservation Unfortunately the initial phase of the Italian government’s support for the highly productive SADC Regional Programme for Rhino Conservation ends in June this year. Plans are afoot to submit a funding proposal to the Italian government for a follow-up project that will build upon the results achieved during the initial phase of support. Rhino status-reporting workshop During the reporting period a successful status-reporting workshop was also held with AfRSG involvement at Kenya Wildlife Service’s training institute in Naivasha. This workshop was part of a Kenyan rhino conservation project sponsored by UK-funded Darwin’s Initiative. AfRSG’s Ben Okita presented the results of his MSc study of the performance of six of the main Kenyan black rhino populations, and this was well received. Progress was made with delegates from various parks in Kenya working on their individual population status reports at the workshop. AfRSG members got an opportunity to look critically at rhino identification master files from different rhino areas and make constructive recommendations on how these could be improved. AfSRG-accredited Kenya Wildlife Service rhino-monitoring instructors Pachyderm No. 38 January–June 2005 distribuée aux délégués et le responsable scientifique du GSRAf a donné une formation rapide à son utilisation, assisté par Sandra Snelling et Ken Maggs, du Parc. Des délégués des 17 agences différentes (y compris le bureau sous-régional d’Interpol), venus de sept pays, ont assisté au cours. Avec le financement du PRCR/ SADC, la formation « Sur les lieux du crime » s’est poursuivie, avec un cours donné en Zambie et d’autres qui sont prévus en Tanzanie et en en Afrique du Sud en mai 2005. Kaziranga fête ses 100 ans Le responsable scientifique du GSRAf a assisté aux célébrations du centenaire du Parc National de Kaziranga en Assam, Inde. Il a donné une présentation lors d’un atelier sur la gestion des prairies dans le parc, atelier qui faisait partie des célébrations du centenaire. Sa présentation s’inspirait de l’expérience africaine pour promouvoir le bien-fondé de la gestion biologique active pour assurer une croissance rapide du nombre de rhinos, bénéfique pour une viabilité accrue à long terme et aussi pour fournir une plus grande et plus stratégique zone tampon contre le braconnage. Ceci a entraîné les débats en séance plénière. On espère que sa visite sera le point de départ d’un partage accru des connaissances et des idées sur la gestion biologique des rhinos entre l’Afrique et l’Asie au bénéfice de la conservation des rhinos dans le futur. Programme régional de la SADC pour la conservation des rhinos Hélas, la première phase du support du Gouvernement italien au très productif Programme régional pour la Conservation des rhinos se termine en juin prochain. On prépare des plans pour soumettre au Gouvernement italien une proposition de financement pour un projet de suivi qui sera basé sur les résultats obtenus au cours de la première phase du support. Atelier de rapport sur le statut des rhinos Durant cette période, il y eut aussi un atelier très constructif sur le rapport du statut avec l’implication du GSRAf, à l’Institut de formation du Kenya Wildlife Service, à Naivasha. Cet atelier faisait partie d’un projet 13 Brooks (trained using the acclaimed AfRSG monitoring training course based on individual identification) have themselves trained staff in rhino-monitoring techniques in a newly restocked area. The encouraging result was that staff created excellent master files without the need for any outside assistance. Browse-availability assessment The rapid browse-availability assessment procedure devised by AfRSG’s Keryn Adcock (as part of the SADC Rhino Management Group’s black rhino carrying-capacity assessment project) was used during the reporting period in assessing the carrying capacities and suitability of five potential black rhino introduction sites in the Serengeti ecosystem during the reporting period, as well as in nine black rhino areas in Kenya. The results of this work provided useful background for a multistakeholder workshop held in January 2005 to discuss future black rhino conservation in the Serengeti ecosystem. Keryn Adcock has also just recently presented a course to selected Ezemvelo–KZN–Wildlife and private sector conservation staff on how to assess browse availability in KwaZulu-Natal, South Africa. Positive review for AfRSG AfRSG’s main sponsor over the last decade, the World Wide Fund for Nature (WWF), commissioned an independent review of its support to AfRSG. The reviewing consultant concluded that one of the most important lessons to be learned was that supporting the basic functions of an institution such as AfRSG (provided it is conservation oriented and has good leadership) and sustaining this support over a number of years can prove to be an extremely cost-effective conservation measure. This flies in the face of the conventional wisdom that holds that core support for an institution is invariably a less effective use of money than direct field action. Despite these comments, raising sufficient funds to maintain a part-time scientific officer and operate effectively remains a challenge. kenyan de conservation des rhinos sponsorisé par l’Initiative de Darwin, financée par le Gouvernement britannique. Ben Okita, du GSRAf, a présenté les résultats du travail qu’il a réalisé pour son MSc sur la performance de six des principales populations kenyanes de rhinos noirs et ceci a été très apprécié. On a fait des progrès avec les délégués venus de divers parcs kenyans qui travaillaient sur leurs rapports individuels sur le statut des populations de rhinos lors de l’atelier. Les membres du GSRAf ont eu l’occasion de jeter un œil critique sur les dossiers d’identification des rhinos venus de différentes aires de rhinos et de faire des recommandations constructives sur la manière de les améliorer. Les instructeurs en surveillance des rhinos du Kenya Wildlife Service accrédités par le GSRAf (formés au moyen du cours de formation très apprécié du GSRAf, basé sur l’identification individuelle) ont eux-mêmes formé du personnel aux techniques de surveillance des rhinos dans une zone qui vient d’être repeuplée. Il est très encourageant de constater que le staff local a créé d’excellents fichiers sans avoir besoin d’aucune aide extérieure. Evaluation de la disponibilité de nourriture La procédure rapide d’évaluation de la disponibilité de nourriture conçue par Keryn Adcock du GSRAf (fait partie du projet d’évaluation de la capacité de charge en rhinos noirs du Groupe de gestion des rhinos de la SADC) a servi pendant la période couverte par ce rapport pour évaluer la capacité de charge et les qualités et les défauts de cinq sites où pourrait avoir lieu l’introduction de cinq rhinos noirs, dans l’écosystème du Serengeti, et dans neuf zones à rhinos noirs situées au Kenya. Les résultats de ce travail ont fourni de très utiles informations préalables à un atelier rassemblant en janvier 2005 de nombreuses parties prenantes pour discuter de la future conservation des rhinos noirs dans l’écosystème du Serengeti. Keryn Adcock vient justement de présenter un cours à un personnel de la conservation choisi d’Ezemvelo-KZN-Wildlife et du secteur privé, sur la façon d’évaluer la disponibilité en nourriture au KwaZulu-Natal, en Afrique du Sud, en se servant de sa méthode. Transition In closing, I sadly inform you of the deaths of two people—Hans Hansen and AfRSG member Mike 14 Pachyderm No. 38 January–June 2005 African Rhino Specialist Group report African Elephant Hearn—who made significant contributions to rhino conservation in Africa. Mike will be remembered for his work with Namibia’s Save the Rhino Trust in conserving the desert black rhino in Kunene. Mike assisted with monitoring (especially managing the database) while undertaking research, as well as through his keen promotion of community-based conservation in this region. The massive effect that Mike had as both a conservationist and a well-liked person was reflected by the many and varied tributes to him on the Save the Rhino International website. Hans will be remembered for his fund raising for rhino conservation in his native Denmark, and in particular for the individual-based photographic rhino monitoring work he undertook in Pilanesberg National Park, South Africa, over a long period, which contributed significantly to this population being one of the best understood in Africa, as well as for a survey he undertook of black rhinos in Aberdares National Park, Kenya. Both will be sadly missed. Révision positive pour le GSRAf Le principal sponsor du GSRAf pour la dernière décennie, le Fonds Mondial pour la Nature (WWF) a demandé une révision indépendante de son support au GSRAf. Le consultant qui a fait ce bilan a conclu que l’une des leçons les plus importantes à retenir était que le fait de soutenir les fonctions de base d’une institution telle que le GSRAf (pour autant qu’elle soit orientée vers la conservation et qu’elle soit bien dirigée) et de maintenir ce soutien pendant des années peut se révéler une mesure de conservation extrêmement rentable. Ceci bat en brèche la sagesse populaire qui prétend que le soutien de base d’une institution est invariablement moins rentable que l’action sur le terrain. Malgré ces commentaires, récolter assez de fonds pour conserver un conseiller scientifique à temps partiel et fonctionner efficacement reste un challenge. Transition Pour terminer, je dois hélas vous faire part du décès de deux personnes–le membre du GSRAf, Mike Hearn, et Hans Hansen—qui ont significativement contribué à la conservation des rhinos en Afrique. On se souviendra du travail de Mike pour le Save the Rhino Trust Namibie, pour la conservation du rhino noir du désert à Kunene. Mike a participé au monitoring (et spécialement à la gestion de la base de données) tout en entreprenant des recherches et il a promu avec enthousiasme la conservation communautaire dans cette région. L’effet considérable qu’il a eu en tant que défenseur de l’environnement et personne très appréciée, a été bien reflété par les hommages nombreux et divers qui sont parvenus sur le site de Save the Rhino International. On se souviendra de Hans pour la récolte de fonds qu’il a menée dans son Danemark natal, et particulièrement pour le travail de surveillance continue des rhinos basée sur des photos individuelles qu’il a réalisé au Parc National de Pilanesberg, en Afrique du Sud, pendant longtemps, et qui a significativement contribué à ce que cette population soit l’une des mieux comprises en Afrique. On se rappellera aussi l’étude qu’il a entreprise sur les rhinos noirs du Parc National des Aberdares, au Kenya. Tous deux nous manqueront beaucoup. Pachyderm No. 38 January–June 2005 15 Asian Rhino Specialist Group report Rapport du Groupe Spécialiste des Rhinos d’Asie Mohd Khan bin Momin Khan, Chair/Président, with/avec Thomas J. Foose and Nico van Strien Programme Officers/Responsables de Programme Malaysian Rhino Foundation, Suite B-6-12, Megan Ave. II, 12 Jalan Yap Kwan Seng, 50450 Kuala Lumpur, Malaysia; email: [email protected] Kaziranga celebrates 100 years Kaziranga fête ses 100 ans Kaziranga National Park celebrated its centenary, 14– 17 February 2005. Established in 1905 as a reserve for the last 10–20 Indian rhinos, Rhinoceros unicornis, believed to survive in Assam, Kaziranga became a national park in 1974 after being gazetted as a number of other types of protected areas during the intervening years. Whatever the official status, Kaziranga’s success as a protected area has been spectacular. From the low point of 10–20 rhinos in 1905, the population under very strict protection from the Forest Department of Assam has grown to at least 1552 in the last full census in 1999 and to an even higher number today. A full census was scheduled for April 2005 (counts are conducted every 6 years) but a combination of early rains and problems with properly preparing the area with controlled burning forced the census to be postponed to 2006. Nevertheless, the rhinos are prospering, as are other megafauna—elephants, buffaloes, tigers, and barasingha or swamp deer. The anniversary celebration convened conservationists to develop a strategy for sustaining and extending Kaziranga’s success into this century. Despite past success, many challenges confront the park and its environs including the very important KarbiAnglong Hills, which provide important seasonal habitat for much of the Kaziranga megafauna. An expanding and developing human population continues to apply pressure on Kaziranga. Major concerns include possible expansion of the trans-Assam highway (which forms the southern boundary of the park and separates it from Karbi-Anglong Hills) and intensifying floods due to increased deforestation upstream. Moreover, poachers are still a threat, which must constantly be counteracted with great diligence and dedication by park rangers. A number of work- Le Parc National de Kaziranga a célébré son centenaire du 14 au 17 février 2005. Institué en réserve en 1905 pour les 10 à 20 derniers Rhinos unicornes (Rhinoceros unicornis) qui, pensait-on, survivaient en Assam, Kaziranga est devenu un parc national en 1974 après avoir été enregistré comme un certain nombre d’autres types d’aires protégées pendant les années intermédiaires. Quel qu’en fût le statut officiel, le succès de Kaziranga en tant qu’aire protégée fut spectaculaire. Des 10 à 20 rhinos estimés en 1905, la population, sous la très rigoureuse protection du Département des Forêts d’Assam a crû jusqu’à un nombre minimum de 1552, relevé lors du dernier recensement complet en 1999 ; il est peut-être encore supérieur aujourd’hui. Un recensement complet devait avoir lieu début 2005 (les comptages se font tous les six ans), mais la combinaison de pluies précoces et de problèmes dans la préparation de la zone par des feux contrôlés a reporté le recensement au début de 2006. Néanmoins, les rhinos prospèrent, comme le reste de la grande faune – éléphants, buffles, tigres et barasinghas ou cerfs des marais. La célébration de cet anniversaire a rassemblé des protecteurs de l’environnement pour qu’ils développent une stratégie afin de soutenir et de prolonger la réussite de Kaziranga dans ce siècle-ci. Malgré ses succès passés, le parc et ses environs font face à de nombreux défis, comme celui que connaissent les très importantes Collines de Karbi-Anglong qui fournissent un habitat saisonnier vital pour une grande partie de la grande faune de Kaziranga. Une population humaine continue à se développer et à s’étendre en exerçant une pression sur le parc. On redoute beaucoup l’extension possible de la grande route trans-Assam (qui forme la limite sud du parc et le sépare des Collines de Karbi-Anglong) et des 16 Pachyderm No. 38 January–June 2005 Asian Rhino Specialist Group report African Elephant shops were conducted, action plans formulated, a Kaziranga Charter adopted and a Friends of Kaziranga Forum established to ensure continued participation by the many stakeholders interested and involved in conserving this magnificent area. Drs Nico van Strien and Tom Foose represented the Asian Rhino Specialist Group. Also participating was Dr Richard Emslie, Scientific Officer of the African Rhino Specialist Group, who was recruited by AsRSG and the WWF AREAS Program to provide perspective for future rhino management in Kaziranga and Assam based on the similarly great success in rhino conservation achieved in southern Africa for the white rhino (Ceratotherium simum simum). A major possibility under consideration is to translocate rhinos from Kaziranga to other protected areas in Assam in places where the species formerly occurred (for example in Manas) but has been eliminated or greatly reduced in numbers, as well as to some new areas with appropriate rhino habitat. Catastrophic decline of the rhino in Nepal While the rhino census in Kaziranga had to be postponed for a year, the Rhino Count 2005 did proceed in April in Chitwan National Park, Nepal, with very unsettling results. The number of Rhinoceros unicornis in Chitwan has declined from the 544 observed in the last full census in 2000 to 372 in the just-completed 2005 count. This catastrophic decline is believed to be due mostly to increased poaching related to political instability in Nepal. However, there are habitat degradation issues as well. Government agencies and NGOs active in rhino conservation in Nepal are intensively reassessing the situation. Global Management and Propagation Board for the Sumatran rhino At a workshop in Jakarta, Indonesia, 20–21 March 2005, various stakeholders from range and non-range states for the Sumatran rhino, Dicerorhinus sumatrensis, convened to improve the management and propagation of individuals of this species maintained in captivity and semi-captivity. A Global Management and Propagation Board was formed, and a number of recommendations were formulated to increase cooperation and coordination. These include Pachyderm No. 38 January–June 2005 inondations croissantes du fait de la déforestation accrue en amont. De plus, les braconniers constituent toujours une menace qui doit sans cesse être contrée avec beaucoup de diligence et de dévouement par les gardes du parc. Il y eut un certain nombre d’ateliers, des plans d’action ont été rédigés, on a adopté une Charte pour Kaziranga et on a créé un Forum des Amis de Kaziranga pour assurer la participation suivie des nombreuses parties prenantes intéressées et impliquées dans la conservation de cette région magnifique. Les Dr Nico van Strien et Tom Foose représentaient le Groupe Spécialiste des Rhinos d’Asie. Il y avait aussi le Dr Richard Emslie, responsable scientifique du Groupe Spécialiste des Rhinos d’Afrique qui avait été recruté par le GSRAs et par le Programme AREAS du WWF pour présenter les perspectives de la gestion future des rhinos à Kaziranga et en Assam, en se basant sur la grande réussite, comparable, que connaît la conservation des rhinos blancs (Ceratotherium simum simum) en Afrique australe. On est en train de penser à transférer des rhinos de Kaziranga vers d’autres aires protégées en Assam, là où l’espèce a vécu auparavant (par exemple à Manas) mais a été éliminée ou fortement réduite, ou encore vers d’autres régions nouvelles qui conviendraient aux rhinos. Déclin catastrophique des rhinos au Népal Si le recensement de Kaziranga a dû être reporté d’un an, le Comptage des Rhinos 2005 a eu lieu en avril au Parc National de Chitwan, au Népal, avec des résultats très inquiétants. Le nombre de Rhinoceros unicornis a baissé à Chitwan des 544 observés lors du dernier recensement complet, en 2000, à 372 lors du comptage qui vient de se terminer. On croit que ce déclin catastrophique est dû principalement au braconnage accrû lié à l’instabilité politique du Népal. On constate aussi, hélas, des problèmes de dégradation de l’habitat. Les organismes gouvernementaux et les ONG actives dans la conservation des rhinos au Népal s’occupent intensément de la réévaluation de la situation. 17 Khan et al. the possibility of relocating animals among existing facilities and better dissemination to all facilities maintaining this species under intensive management of information on methods that have proven successful in reproducing it, especially those employed at the Cincinnati Zoo. 18 Conseil pour la Gestion Globale et la Propagation du Rhino de Sumatra Lors d’un atelier à Jakarta, en Indonésie, les 20 et 21 mars 2005, diverses parties prenantes venues, ou non, des états de l’aire de répartition du rhino de Sumatra (Dicerorhinus sumatrensis) ont convenu d’améliorer la gestion et la propagation des individus de cette espèce qui sont gardés en captivité et en semicaptivité. Un Conseil pour la Gestion Globale et la Propagation des Rhinos fut créé, et de nombreuses recommandations furent émises pour améliorer la coopération et la coordination. Elles incluent la possibilité de déplacer des animaux entre les installations existantes, et une meilleure distribution vers toutes les installations qui conservent cette espèce en gestion intensive des informations sur les méthodes qui ont permis la reproduction, et spécialement celles utilisées au Zoo de Cincinnati. Pachyderm No. 38 January–June 2005 Changes in elephant numbers in eastern and southern Africa RESEARCH Changes in elephant numbers in major savanna populations in eastern and southern Africa J. Julian Blanc,1* Richard F.W. Barnes,2 G. Colin Craig,3 Iain Douglas-Hamilton,4 Holly T. Dublin,1,5 John A. Hart 6 and Chris R. Thouless7 1 IUCN/SSC African Elephant Specialist Group, PO Box 68200, Nairobi 00200, Kenya; email: [email protected]; *corresponding author 2 Division of Biological Sciences 0116, University of California at San Diego, La Jolla, CA 92093-0116, USA; Environmental Sciences Research Centre, Anglia Polytechnic University, Cambridge CB1 1PT, UK 3 PO Box 25476, Windhoek, Namibia 4 Save the Elephants, PO Box 54667, Nairobi 00200, Kenya 5 IUCN Species Survival Commission (SSC), c/o South African National Biodiversity Institute, Rhodes Drive, Claremont, Cape Town 7735, South Africa 6 Wildlife Conservation Society, DRCongo Program, Kinshasa 7 PO Box 209, Timau, Kenya Abstract This paper presents an analysis of changes in elephant population estimates selected from the two most recent reports of the African Elephant Database (AED). Sites selected for analysis were restricted to surveyed areas in which successive estimates had been made using comparable methods. The resulting selection consisted of surveys conducted in eastern and southern Africa between 1994 and 2002, which together cover a large percentage of the total elephant population for which estimates are available in these two regions. The results suggest a significant overall increase (p < 0.0002) for the eastern and southern African sites combined. The overall increase in the southern African estimate was significant (p < 0.0004), but the increase in eastern African estimates was not statistically significant. It is concluded that savanna elephant populations in eastern and southern Africa are more likely to have increased than to have declined in the years leading up to the African Elephant Status Report 2002; important caveats become evident, however, when interpreting these findings. Résumé Cet article analyse les changements dans les estimations de populations d’éléphants selon les plus récents rapports de la Base de Données de l’Eléphant africain (BDEA). Les sites choisis pour cette analyse ont été limités aux aires étudiées dans lesquelles on a fait des estimations successives en utilisant des méthodes comparables. Les résultats de la sélection reprennent des études menées en Afrique orientale et australe entre 1994 et 2002 qui, ensemble, couvrent un grand pourcentage de la population d’éléphants pour laquelle des estimations existent dans ces deux régions. Les résultats suggèrent une augmentation générale significative (p < 0,0002) pour les deux sites combinés. L’augmentation globale pour les estimations en Afrique australe était significative (p < 0,0004), mais l’augmentation des estimations en Afrique orientale n’était pas statistiquement significative. On en conclut que les populations d’éléphants de savane en Afrique orientale et australe sont plus susceptibles d’avoir augmenté que diminué durant les années qui ont précédé le Rapport 2002 sur le Statut de l’Eléphant africain ; il reste d’importantes mises en garde, cependant, quant à l’interprétation de ces découvertes. Pachyderm No. 38 January–June 2005 19 Blanc et al. Introduction A question of major interest to decision-makers involved in the conservation and management of African elephants is whether the elephant population on the continent as a whole is increasing or decreasing. Since its inception in 1986, the African Elephant Database (AED), probably the most comprehensive effort to monitor the distribution and abundance of any widely distributed species of mammal in the wild, has been periodically reporting on the status of African elephant populations throughout the species’ range. Data for the AED are obtained from a wide variety of sources ranging from systematic aerial counts to ad hoc guesses, and the reliability of the estimates varies accordingly. To characterize this variability, the AED separates elephant numbers into four categories of certainty—definite, probable, possible and speculative—according to fixed rules (see Blanc et al. 2003 for details). In this system, if all elephant populations were systematically surveyed, giving unbiased estimates with measured precision, the sum of the definite and probable categories would be an accurate statement of true elephant numbers. The reported numbers under the definite and probable categories in the continental, all regional and most national sections in the African Elephant Status Report (AESR) 2002 (Blanc et al. 2003) are higher than the corresponding numbers reported in the African Elephant Database 1998 (Barnes et al. 1999). These changes, however, do not necessarily reflect real overall increases in elephant numbers, and a casual comparison of these figures is likely to be misleading for a number of reasons. Many of the continent’s elephant populations have never been systematically surveyed. Most elephant surveys tend to concentrate in and around protected areas, although up to 80% of elephant range may lie outside them. Any changes reported are only derived from a subset of all elephant populations and may therefore not reflect overall changes in numbers. The extent of unsurveyed range across the continent may amount to as much as 50% of total elephant range in Africa (Blanc et al. 2003), but even this estimate is subject to considerable uncertainty. Elephant distribution data for the AED are obtained from questionnaire replies and other potentially unreliable sources, which can quickly become outdated. To quantify this uncertainty, the AED has recently begun to categorize elephant range into three categories of reliability: known range, possible range and doubtful range. 20 Many important populations are surveyed infrequently or have been surveyed only once. In consequence, any one report of the AED may repeat some estimates from the previous report because these are still the most up to date available. This makes using total numbers invalid as a measure of change, as constancy of numbers at some sites reflects only the same information carried forward from one status report to the next. The totals in the definite and probable categories may decline where an out-of-date estimate has been degraded to the speculative category and no more recent information is available. Conversely, where a population is surveyed for the first time, the resultant increase in the total is due not to population increase but to the inclusion of new information. False increases (or decreases) may also happen when the boundary of the study area changes between surveys, although the site name remains the same. When only parts of the ranges of elephant populations are included in the surveyed area, changes in estimates may be caused by elephant movements rather than real changes in population size. Even where two successive surveys of the same area are available, misleading changes may be observed when different methods, liable to different levels of accuracy or bias, are used in the two consecutive surveys. Variation in survey conditions—like the time of the year or even the use of different survey crews— may result in changes in numbers of elephants seen, thus contributing to differences recorded over time. In addition many estimates come from sample surveys and are therefore subject to statistical sampling error. As a result, differences between successive estimates could be due purely to chance but still make a large contribution to the differences between totals. Despite these problems, it is possible to select those sites where surveys have been repeated using comparable methods and to conduct a formal comparison of elephant numbers over time restricted to a segment of the continental population. As it turns out, a large proportion of the known elephant populations in eastern and southern Africa can be included in such a sample. This paper presents the results of such an analysis using data taken from the AED 1998 and the AESR 2002. Methods Site selection Survey data were obtained from the two most recent reports of the AED (Barnes et al. 1999; Blanc et al. Pachyderm No. 38 January–June 2005 Changes in elephant numbers in eastern and southern Africa 2003). Data from previous reports of the African Elephant Database (such as Said et al. 1995) were not used in this analysis, as there were few sites with comparable data across all three reports. Surveys were regarded as comparable if they met the following conditions: • Similar survey methods or methods with similar levels of accuracy were used in both time periods. All guesses were excluded. • Approximately the same elephant range was covered in both surveys. However, where a discrete population of elephants was censused in its entirety in both surveys, the estimates were regarded as comparable even if the total areas were different. In some cases, adjacent input zones reported in the AED 1998 were combined to match larger input zones reported in the AESR 2002. Although some total ground counts and individual recognition studies met the above criteria, the majority of the sites in the sample were covered by aerial surveys. Systematic aerial surveys fall into two broad categories: sample counts and total counts. In the former, a representative sample of the study area is usually covered by counting animals along transects of known width on either side of flight lines. The overall density of elephants recorded in the transects is used to calculate a population estimate with confidence limits for the entire study area. The width of the confidence interval depends on the number and distribution of animals in the study area as well as on the sampling design and intensity (Norton-Griffiths 1978). Aerial total counts, on the other hand, aim to record every elephant in a study area by flying closely spaced flight lines to cover the entire area. Since it is assumed that every animal is counted, aerial total counts give no estimates of precision. However, as some animals are always likely to be missed, total counts tend to result in undercounts and as such represent a minimum estimate of the true number of animals present. Nevertheless, for the purposes of this analysis, all total counts were treated as sample counts with zero variance (Norton-Griffiths 1978). Two important sites in eastern Africa were excluded from the analysis, as their inclusion would have biased the overall results. The Ruaha–Rungwa ecosystem in Tanzania with an estimated 24,682 ± 6495 (estimate ± 1.96 H SE) elephants in 2002 was excluded because the 1998 estimate (13,021 ± 4300) is believed to have been an undercount (Conservation Information and Monitoring Unit 2002). Similarly, the dif- Pachyderm No. 38 January–June 2005 ference between the 1998 and 2002 dung count survey estimates for Mount Kenya (4022 ± 1083 and 2911 ± 640 respectively), is not due to changes in elephant density but rather to an improvement in the estimate of the area available to elephants in the ecosystem (H. Vanleeuwe, pers. comm.) One of the sites in eastern Africa, Samburu District (Kenya), was included in the sample despite an increase in precision due to a change in survey method between the AED 1998 (aerial sample count) and the AESR 2002 (aerial total count). However, since total counts tend to result in undercounts the inclusion of Samburu District in the analysis is justified on the basis that the 2002 total count estimate is larger than the upper 95% confidence limit of the 1998 sample count. The above criteria resulted in a selection of 51 sites in eastern (13 sites; table 1) and southern Africa (38 sites; table 2). The geographical location of the selected sites is shown in figure 1. A small number of comparable surveys also existed in Central and West Africa (1 site in Central Africa—Garamba National Park—and 7 sites in West Africa, 5 of which are part of the same population in Burkina Faso). However, these sites were excluded from the analysis as they represented very small fractions of total known plus possible range in eastern Africa (0.2%) and in southern Africa (2%). The analysis reported here is, in consequence, restricted to sites in eastern and southern Africa. The combined area covered by the selected surveys represents 21% of total known plus possible for eastern Africa and 23% for southern Africa, but only 11% of the total current known and possible elephant range estimate for the continent. The sites include some of the best-managed and best-studied parks on the continent and cannot therefore be considered a representative sample of sites with elephants in Africa as a whole. However, the total number of elephants in these sites (353,687) represents a high proportion (77%) of total continental numbers under the definite and probable categories reported in the AESR 2002 (461,091). Differences and rates of change Estimates were added separately for each dataset and differences between the two datasets were calculated at the levels of individual and combined regions. The statistical significance of these differences was judged by calculating the variance of the estimated difference and constructing a 95% confidence interval for the true difference. A difference was deemed significant if its 21 Blanc et al. Ethiopia Sudan Cameroon Somalia regional boundary 2 selected sites 6 Uganda 3 Elephant Congo range 5 Gabon known Kenya 4 12 Democratic Republic of Congo possible doubtful 1 7 Tanzania 13 8 9 Angola 44 46 i Zambia 38 w Mala 42 43 41 45 39 40 10 11 17 Mozambique 51 16 48 50 18 19 20 14 49 Zimbabwe Namibia 47 15 Botswana 27 28 26 31 37 2436 34 29 22 32 23 South Africa 35 33 30 21 25 0 300 600 1200 km Figure 1. Location of the sites included in the analysis and elephant range from Blanc et al. (2003). Numbers refer to sites listed in tables 1 and 2. 22 Pachyderm No. 38 January–June 2005 Changes in elephant numbers in eastern and southern Africa Table 1. Eastern African sites used in this study and summary survey tween individual pairs is compared results against the pooled variance of the Survey zone Year of survey Kenya 1. Amboseli 1998 2002 2. Kerio Valley 1997 2002 3. Laikipia 1996 2002 4. Masai Mara 1998 2002 5. Meru and Bisanadi 1997 2002 6. Samburu District 1996 2002 7. Tsavo 1994 2002 Tanzania 8. Katavi–Rukwa 9. Kilombero 10. Mikumi 11. Selous 12. Serengeti 13. Ugalla River 1995 2002 1994 2002 1994 2002 1994 2002 1998 2000 1996 1999 Total eastern sites 1996* 2002* Difference Survey Estimate method 95% CL (±) Area (km2) IR1 IR1 AT3 AT3 AT3 AT3 AT3 AT3 AT3 AT2 AS2 AT2 AT3 AT3 980 1,100 652 490 2,436 3,241 1,450 2,116 360 372 1,224 2,206 7,371 9,221 0 0 0 0 0 0 0 0 0 0 898 0 0 0 5,547 5,547 4,616 4,616 7,000 8,406 3,488 3,488 2,849 2,053 21,095 20,073 38,300 37,382 AS2 AS3 AS2 AS3 AS2 AS3 AS2 AS3 AT3 AT3 AS2 AS3 4,998 5,751 1,903 6,203 700 1,144 49,571 57,886 2,015 1,631 761 1,177 2,360 4,549 514 4,639 309 923 11,025 14,518 0 0 655 615 13,341 11,862 6,928 6,006 3,215 3,069 81,838 80,287 16,860 16,860 6,524 7,252 – – 74,421 92,538 11,345 15,944 211,601 206,901 18,117 19,569 –4,700 6* Survey method codes: IR – individual registration; GT – ground total count; AT – aerial total count; AS – aerial sample count. The number that accompanies the survey method code gives an indication of survey quality, ranging from 1 to 3 (best to worst). For more details, see Blanc et al. (2003). The 95% CL (±) column denotes the standard error of the estimate times 1.96. * median value 95% confidence interval did not overlap zero. In particular, an increase was considered significant if the lower bound of the 95% confidence interval was greater than zero. To calculate p values, a t-statistic was calculated as the ratio of the estimated difference to its standard error. This approach is similar to a matched-pairs t-test in that the sum of the differences be- Pachyderm No. 38 January–June 2005 dataset, but the measure of variation used in this case originates from the sum of the internal variances of sample counts and not from the variance of differences across sites. To express the difference as a rate of change, the finite rate of the increase λ (Caughley 1977) was estimated from the ratio λt of the pooled estimates from the AED 1998 ( ) and the AESR 2002 ( ): (Eq. 1) where t is the time interval in years. As the time interval between surveys in the AED 1998 dataset and the AESR 2002 dataset varied from one site to another (see under Results below), it was adjusted by calculating t as the average time interval between surveys weighted by the number of elephants in the AED 1998 dataset. The finite rate of increase λ is therefore estimated by (Eq. 2) From this the estimated mean an can be nual per cent rate of increase calculated as (Eq. 3) To construct a 95% confidence interval around the estimated rates of change, the variance of the ratio of population estimates was approximated using a second-order Taylor series expansion (Schreuder et al. 2004): (Eq. 4) 23 Blanc et al. Table 2. Southern African sites used in this study and summary survey results Survey zone Botswana 14. Northern Botswana 15. Tuli Mozambique 16. Magoe South 17. Niassa Namibia 18. Caprivi 19. Etosha 20. Kaudom/Nyae Nyae South Africa 21. Addo Elephant 22. Atherstone 23. Itala 24. Klaserie 25. Knysna 26. Kruger 27. Madikwe 28. Makuya 29. Marakele 30. Mkuzi 31. Phalaborwa 32. Pilanesberg 33. Pongola 34. Sabie Sand 35. Tembe Elephant 24 Year of Survey Estimate survey method 95% CL (±) where Var is the variance and Cov the covariance. In this analysis, however, the errors in the estimates are uncorrelated and the covariance can therefore be taken to be zero. Area (km2) 1995 1999 1994 2001 AS3 AS2 AS1 AT2 89,227 120,604 831 1,262 13,406 122,922 21,237 146,050 456 885 0 885 1995 2001 1998 2002 AS2 AS2 AS2 AS2 137 1,264 8,707 13,061 187 1,359 1,937 2,433 2,824 2,824 42,349 42,341 1995 1998 1995 2000 1995 2000 AS2 AS2 AS1 AS2 AS2 AS2 4,883 4,576 1,189 2,100 1,085 1,966 1,247 1,223 410 774 545 973 19,290 18,259 22,270 19,269 15,020 12,107 1998 2002 1997 2002 1997 2002 1997 2002 1997 2002 1998 2002 1998 2002 1997 2002 1998 2002 1997 2002 1997 2002 1997 2002 1997 2002 1997 2002 1997 2002 AT1 AT1 AT3 AT3 GT1 GT1 AT3 AT2 IR1 IR1 AT3 AT2 AT3 AT3 AT3 AT2 IR1 IR1 IR1 IR1 AT3 AT2 AT3 AT3 IR1 IR1 AT3 AT2 AT3 AT3 272 337 24 32 45 61 303 467 3 4 8,869 10,459 282 318 8 27 48 91 25 28 42 23 87 142 20 33 311 757 115 140 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 513 513 136 136 297 297 628 628 300 300 19,624 19,624 700 700 165 165 380 380 380 380 41 41 553 553 119 119 572 572 300 300 Results Table 3 presents a summary of changes in elephant numbers, while table 4 summarizes the rates of change. Results are discussed in detail below, first for both regions combined and then for each region in turn. Southern and eastern African sites combined Surveys selected from the AED 1998 were conducted between 1994 and 1998; the median survey year was 1997. For surveys selected from the AESR 2002, survey years ranged from 1998 to 2002, the median year being 2002. The time difference between surveys ranged from a minimum of 2 years to a maximum of 8, with a median of 5 years (25% quartile: 4 years; 75% quartile: 5 years). The weighted mean time between surveys was 5.097 years. The total area surveyed increased slightly from the AED 1998 to the AESR 2002 by 37 km2, or 0.006%. A significant proportion of the selected sites (41 out of 51, or 80.4%) reported higher estimates in the AESR 2002 dataset than in the AED 1998 dataset (sign test, Z = 4.20, p < 0.0001). The estimated total number of elephants in the tabulated sites increased from 282,895 in the 1998 dataset to 353,687 in the 2002 dataset, a difference of 70,792 or 25%. The standard error of the difference was 14,464. Taking t to be 1.96, this gives a 95% confidence interval of +35,863 to +141,584 for the overall difference in the tabulated Pachyderm No. 38 January–June 2005 Changes in elephant numbers in eastern and southern Africa Table 2. (continued) Survey zone 36. Timbavati Year of Survey Estimate 95% survey method CL (±) Area (km2) 1997 2002 1997 2002 AT3 AT2 AT3 AT2 322 372 134 88 0 0 0 0 784 494 144 144 1997 2001 39. Luambe and Lumimba 1996 1999 40. Lupande 1996 2002 41. Munyamadzi 1996 2000 42. Musalangu 1996 2001 43. North Luangwa 1996 2000 44. Sandwe and Chisomo 1996 1999 45 South Luangwa 1996 2002 46. West Petauke 1996 1999 AS2 AS3 AS3 AS2 AS2 AS2 AS2 AS3 AS3 AS3 AS3 AS3 AS2 AS2 AS2 AS2 AS3 AS2 4,482 2,194 763 1,053 892 975 102 1,108 305 1,121 3,033 3,750 818 128 7,942 4,459 2,435 897 3,222 5,590 811 551 1,394 587 210 500 690 898 2,252 1,076 1,597 155 2,930 1,519 2,773 1,399 22,400 16,929 4,462 4,462 4,840 4,959 3,300 3,292 17,350 17,350 4,636 4,636 4,920 750 9,050 8,448 4,140 905 37. Umbabat Zambia 38. Kafue To examine whether the significant difference was simply a consequence of increases in a few very large populations, the analysis was repeated excluding sites with very large increases in absolute numbers. The sites excluded were northern Bo- tswana (with a difference of +31,377), northwest Matabeleland (+13,030) and Selous (+8,315). The results of the analysis on this reduced dataset still show a significant difference of between +5,137 and +31,003 elephants (t = 2.74, p < 0.01). In view of this, the three sites were included in subsequent analyses. Eastern Africa Surveys reported in the AED 1998 were conducted between 1994 and 1998, and all but one of the surveys reported in the AESR 2002 were carried out in 2002. The time difference between surveys in the selected sites in eastern Africa ranged from 2 to 8 Zimbabwe 47. Gonarezhou, Malipati 1996 AS2 3,842 1,692 5,435 years, with a median time difference and Mahenye 2001 AS2 4,992 1,577 5,346 of 6 years and a weighted mean time 48. Mavuradonha and 1997 AS1 120 120 617 difference of 7.45 years. The total reGreat Dyke 2001 AS2 13 26 617 ported surface area surveyed in the 49. NW Matabelelanda 1997 AS2 36,280 7,308 25,074 selected sites declined between the 2001 AS2 49,310 6,028 25,072 AED 1998 and the AESR 2002, by 50. Sebungweb 1997 AS2 13,386 1,241 15,597 4700 km2 (–2.22%), but all the sites 2001 AS2 13,989 2,098 15,622 that registered declines in area reported 51. Zambezi Valley 1995 AS2 17,105 2,580 14,842 2001 AS2 18,948 2,463 17,127 higher elephant estimates in the latter. The AESR 2002 reports higher Total southern sites 1997* – 208,474 16,959 387,859 elephant estimates for 11 of the 13 2002* – 261,149 23,441 392,596 selected eastern African sites with an Difference 5* – 52,675 28,933 4,737 estimated difference of 18,117 elephants. While the 95% confidence inSee footnote to table 1 for survey method codes and additional details. terval of the difference (–1452 to a Includes the following survey zones: Hwange, Matabeleland and Matetsi b +36,234) includes the value zero, and Includes the following survey zones: Binga, Chete, Chirisa, Chizarira, Kariba, Lusulu, Matusadona, North Gowke and Sijarira the result is therefore not statistically *median value significant (t = 1.82, p > 0.05), the range of probabilities lies mostly on populations, a statistically significant difference (t = the positive side of zero. This is also true of the esti3.97; p < 0.0002). This translates to an annual rate of mated annual rate of increase (2.97%, range –0.59% to increase of 4.48%, with a 95% confidence interval 5.87%), making an overall increase substantially more ranging from +2.17% to +6.60%. likely than a decline in numbers. Pachyderm No. 38 January–June 2005 25 Blanc et al. Table 3. Results of the analysis of differences in elephant populations in the selected sites Difference Variance of estimate Standard error of difference t Area (km2) Median survey year Estimate Region p Eastern Africa 211,601 206,901 1996 2002 74,421 92,538 18,117 33,505,371 66,174,523 9,984 1.815 >0.05 Southern Africa 387,859 392,596 1997 2002 208,474 261,149 52,675 74,870,543 143,037,079 14,762 3.568 <0.001 Total 599,460 599,497 1997 2002 282,895 353,687 70,792 108,375,914 209,211,602 17,821 3.972 <0.0002 Source: AED 1998; AESR 2002 Table 4. Results of the analysis of rates of change in elephant populations in the selected sites Region Weighted time difference (years) Annual rate of change (%) Variance of ratio Lower 95% CL of rate (%) Upper 95% CL of rate (%) Eastern Africa 7.449 2.97 0.02137 –0.59 5.87 Southern Africa 4.257 5.43 0.00600 2.28 8.31 Total 5.097 4.48 0.00473 2.17 6.60 Source: AED 1998; AESR 2002 Southern Africa The median time difference between the selected surveys in southern Africa was 5 years, with a minimum difference of 3 years and a maximum of 7. The weighted mean time difference between surveys was 4.26 years. The total surveyed area increased by 4737 km2 (+1.22%). Out of 38 sites selected in southern Africa, 30 reported higher estimates in the AESR 2002 with increases ranging from a single elephant (Knysna, South Africa) to over 30,000 (northern Botswana). The remaining 8 sites reported lower estimates in the AESR 2002, ranging from a decrease of 19 elephants in Phalaborwa (South Africa) to 3483 in South Luangwa (Zambia). The total difference between estimates in the AED 1998 and the AESR 2002 for the southern African sites amounts to 52,675 with a 95% confidence interval of +23,742 to +81,608, a highly significant difference (t = 3.57, p < 0.001). The estimated mean annual rate of increase for the selected sites in this region is 5.43% with lower and upper 95% confidence limits of +2.28% and +8.31% respectively. The difference for southern Africa is of sufficient magni- 26 tude to make the combined results of eastern and southern African sites highly significant despite the lack of statistical significance for the eastern African sites alone. Discussion The results of this analysis strongly suggest an overall increase in the number of elephants in the tabulated sites in southern Africa as well as for the combined eastern and southern African sites between 1994 and 2002. Only within-survey variance is accounted for in this analysis, as variance due to changes in survey conditions and movement of elephants across survey boundaries cannot be measured, given only one pair of surveys per site. It is unlikely, however, that allowance for this could have reduced the results to non-significance. It should be emphasized that the results refer only to the relevant total numbers as there are insufficient data in most cases to make meaningful comparisons at the site level. Nevertheless, populations are believed to be increasing in many of the sites listed (see Barnes et al. 1999 and Blanc et al. 2003 for details on individual populations). Pachyderm No. 38 January–June 2005 Changes in elephant numbers in eastern and southern Africa The number of elephants included in the selected sites from the AESR 2002 dataset represents a high proportion (68%) of the definite plus probable elephants for eastern Africa and southern Africa (97%) reported in Blanc et al. (2003), and accounts for virtually the entire elephant populations of some countries: Botswana, Namibia, South Africa and Zimbabwe. The analysis does, however, exclude most of Mozambique; substantial portions of elephant range in Kenya and Tanzania; and the entire ranges of Angola, Sudan and Uganda among others. While the estimated annual rates of increase reported here are similar to those derived from detailed demographic studies (for example, Moss 2001; Wittemyer et al. 2005), and below the theoretical (7%) and observed (10%) long-term maxima (Calef 1988; Foley 2002), it is impossible to determine whether these changes are due solely to natural population growth, or the extent to which they may have been influenced by immigration into survey zones. The selection of sites for this analysis largely comprises protected areas, some of them surrounded by large areas of unsurveyed range. In view of increasing range loss due to habitat conversion in much of the continent (Parker and Graham 1989), net immigration into well-protected areas may be expected and the possibility that elephant movements may have contributed to the observed increases cannot therefore be ruled out. On the other hand, much of the excluded unsurveyed range is unlikely to contain high densities of elephants, and in some southern African countries elephants have expanded into new range in the past few years (Blanc et al. 2003). A large unidirectional bias is therefore unlikely. Despite the limitations inherent in the data, this analysis suggests an overall increase in the number of elephants in southern African sites during the period covered by this comparative study. The likelihood is also high that the eastern African sites experienced an overall increase in elephant numbers during the period. This is a noteworthy finding, as the vast majority of the savanna subspecies of the African elephant are in southern and eastern Africa. It must be reiterated, however, that this says nothing about the situation in West or Central Africa, where data are insufficient to draw any conclusions; more extensive and regular survey work of consistent quality will be required to detect changes in elephant populations in the continent as a whole. Similarly, the results do not imply a uniform increase across all Pachyderm No. 38 January–June 2005 sites but merely an increase on average; some sites may have suffered a decline. Finally, it should not be assumed that elephant populations analysed in this study are continuing to increase at present as the period of the observed increase was centred around the late 1990s. Doubts about the validity of the present results hinge on elephant movements into surveyed areas, the absence of information about the status of elephants in unsurveyed range, and in particular the extent to which the observed increases result from reproduction instead of immigration from unsurveyed areas. Coordinated surveys of entire populations, across international borders where necessary, would remove much of this uncertainty. Clearly, the need continues to obtain reliable estimates for all elephant range. Acknowledgements This paper was completed with financial assistance from the European Union, the US Fish and Wildlife Service, and the UK Department for Environment Food and Rural Affairs. The authors are grateful to Ken Burnham and Bob Burn as well as two anonymous reviewers for invaluable advice and comments. References Barnes RFW, Craig GC, Dublin HT, Overton G, Simons W, Thouless CR. 1999. African Elephant Database 1998. IUCN/SSC African Elephant Specialist Group, Gland, Switzerland. Blanc JJ, Thouless CR, Hart JA, Dublin HT, Douglas-Hamilton I, Craig GC, Barnes RFW. 2003. African elephant status report 2002: an update from the African Elephant Database. IUCN/SSC/ African Elephant Specialist Group, Gland, Switzerland. Calef GW. 1988. Maximum rate of increase in the African elephant. African Journal of Ecology 26:323–327. Caughley G. 1977. Analysis of vertebrate populations. John Wiley and Sons, Chichester. Conservation Information and Monitoring Unit. 2002. Aerial census in the Ruaha–Rungwa ecosystem, dry season, 2002. Unpublished report. Tanzania Wildlife Research Institute, Arusha. Foley C. 2002. High incidence of elephant twin births in Tarangire National Park, Tanzania. Pachyderm 32:64– 66. Moss CJ. 2001. The demography of an African elephant (Loxodonta africana) population in Amboseli, Kenya. Journal of Zoology (London) 255:145–156. 27 Blanc et al. Norton-Griffiths M. 1978. Counting animals. Publication no 1 in series: JJR Grimsdell, HT Russell, eds. Techniques currently used in African wildlife ecology. AWLF/SEMP, Nairobi. Parker ISC, Graham AD. 1989. Men, elephants and competition. Symposium of the Zoological Society of London 61:241–252. Said MY, Chunge RN, Craig GC, Thouless CR, Barnes RFW, Dublin HT. 1995. African Elephant Database 1995. IUCN, Gland, Switzerland. 28 Schreuder HT, Ernst R, Ramirez-Maldonado H. 2004. Statistical techniques for sampling and monitoring natural resources. Forest Service, United States Department of Agriculture. http://www.fs.fed.us/rm/pubs/ rmrs_gtr126.pdf Wittemyer G, Daballen D, Rasmussen H, Kahindi O, Douglas-Hamilton I. 2005. Demographic status of elephants in the Samburu and Buffalo Springs National Reserves, Kenya. African Journal of Ecology 43:44–47. Pachyderm No. 38 January–June 2005 Systematic recording of human–elephant conflict, Tanzania Systematic recording of human–elephant conflict: a case study in south-eastern Tanzania Cyprian Malima,1 Richard Hoare2* and J. Julian Blanc3 1 World Wide Fund for Nature, Tanzania Programme Office, PO Box 63117, Dar es Salaam, Tanzania; email: [email protected] 2 TAWIRI Wildlife Veterinary Programme, PO Box 707, Arusha, Tanzania; email: [email protected]; *corresponding author 3 African Elephant Specialist Group, PO Box 68200, City Square 00200, Nairobi, Kenya; email: [email protected] Abstract A standardized data collection system recommended by the African Elephant Specialist Group to record and assess human–elephant conflict was used in subsistence agricultural areas to the east of the Selous Game Reserve in Tanzania. Nine enumerators were recruited, trained and supervised to collect primary data on elephant damage incidents in 38 rural villages widely spaced throughout an area of 30,000 km2. Losses were suffered over a small total area of cultivation that covered only 1% of potential elephant habitat. In the first year of recording 1239 incidents occurred, of which 973 were assessed as crop raids. Sixteen categories of food crops were damaged, representing loss to both wet- and dry-season produce. Elephants killed two people and people killed 25 elephants. One year of a proposed three-year study already highlights the usefulness and cost-effectiveness of simple, inexpensive recording schemes, operated principally by people within affected communities and producing rapid results relevant to local wildlife management and community-based conservation. Further assessments of the economic value of losses, the selectivity of crops by elephants and tests of causative factors in human–elephant conflict will be made once the data of the three years become available. Résumé On a utilisé un système standardisé de récolte des données recommandé par le Groupe Spécialiste des Eléphants d’Afrique pour rapporter et évaluer les conflits hommes–éléphants dans des zones d’agriculture de subsistance à l’est de la Réserve de Faune de Selous, en Tanzanie. On a recruté neuf compteurs qui ont été formés et supervisés, pour qu’ils récoltent les données de base sur des incidents de dégâts aux cultures dans 38 villages ruraux distribués sur une superficie de 30.000 km2. Les pertes ont été constatées sur une petite surface de culture dont le total ne couvrait que 1 % de l’habitat potentiel des éléphants. Au cours de la première année, on a rapporté 1239 incidents et 973 d’entre eux ont été considérés comme des dégâts aux cultures. 16 catégories de récoltes ont été endommagées, représentant des pertes de produits de saison sèche et des pluies. Les éléphants ont tué deux personnes et les gens ont tué 25 éléphants. Une seule année de cette étude qui devrait durer trois ans a déjà mis en évidence l’utilité et la rentabilité de programmes de rapport simples et peu coûteux, réalisés principalement par des personnes des communautés affectées, qui donnent des résultats rapides intéressant la gestion locale de la faune sauvage et la conservation communautaire. Des évaluations plus poussées de la valeur économique des pertes, de la sélectivité exercée par les éléphants et des tests de facteurs de causalité dans les conflits hommes–éléphants seront faites dès que les données récoltées pendant ces trois ans seront disponibles. Pachyderm No. 38 January–June 2005 29 Malima et al. Introduction Established during the colonial era, the Selous Game Reserve (SGR) in south-eastern Tanzania is the largest protected area (50,000 km2) south of the Sahara under a single management unit. The primary purpose at the time of gazetting the large tract of land was to conserve elephants, which were thought to be fast disappearing (Matzke 1976). In the 1940s some of the sparse human settlements in the area were abandoned due to severe crop damage by elephants, and the vacated land was annexed to SGR (Nicholson 1969). After independence in 1961 the Tanzania government, realizing the earning potential of wildlife populations within the reserve, initiated a programme in which hunting safaris by tourists were given access to the reserve under strictly controlled conditions (Nicholson 1969). In Tanzania, national parks are fully protected, allowing only non-consumptive tourism, but in game reserves limited offtake of trophy animals is permitted through annual quotas. When figures were first produced in 1976 the Selous elephant population was estimated at 110,000—one of the largest on the African continent. Rampant poaching in the 1970s and 1980s, however, reduced the population to about 30,000 by 1989. Today the elephant population in the greater Selous ecosystem (105,000 km2 of surveyed elephant range) is recovering strongly and is estimated to have reached around 70,000 (Blanc et al. 2003). Of these, about 25% are thought to be found outside the various protected areas, but the number fluctuates depending on season and the intensity of the expanding human activity in that area encroaching on elephant habitats and movement routes—agriculture, settlement, hunting, fishing, uncontrolled fires, etc. (Mpanduji et al. 2002). Conflict between humans and wildlife, particularly elephant, hippopotamus, bush pig and wild carnivores, is a continuing serious issue in areas surrounding the SGR. Previous efforts to mitigate elephant damage to crops in the area have largely depended on centralized problem animal control (PAC) units of government-employed wildlife personnel. Due to logistical constraints similar to those experienced elsewhere in Africa (Osborn and Parker 2002) and the recurring nature of the problem, these units have had little lasting effect. Reliance on PAC units has recently diminished due to cutbacks in government spending under the country’s economic adjustment programme. Thus 30 the problem of managing elephants, as is increasingly the case in many African countries, is now de facto largely in the hands of rural communities that interface with elephant range. But local communities and farmers lack the capacity to deal with the problem effectively, so their support for any conservation initiative is jeopardized. In response to the need to evaluate and compare vastly differing human–elephant conflict (HEC) situations across Africa (Dublin and Hoare 2004), some years ago the African Elephant Specialist Group (AfESG) developed a standardized data collection and analysis protocol for HEC situations (Hoare 1999a). This protocol and its associate, a training package for enumerators of elephant damage (Hoare 1999b), were adopted and tested in this study. The project was designed to generate baseline information on HEC levels as such information is an essential first step towards more local management of the conflict problem. This paper describes the performance of the first year of the standardized reporting scheme and discusses benefits of using such a model elsewhere in the African elephant range. Study area The study was carried out around villages in Rufiji, Kilwa and Liwale Districts in south-eastern Tanzania, situated in nine administrative wards: three in each district. These wards border SGR’s eastern zone with the exception of Kikole Ward of Kilwa District, which was chosen because elephants occur locally the year round (fig. 1). The total area covered by the incident-reporting scheme was 30,000 km2, containing 38 villages with an estimated human population of 52,880 people (Tanzania Census 2002). The vegetation is predominantly miombo-dominated (Brachystegia spp.) woodland with undulating topography. The area receives rainfall biannually with an average annual total of 800 mm. ‘Short rains’ fall in November and December and ‘long rains’ may last from February to May. Human communities are primarily artisanal peasants and hunters. Most local people continue to carry out shifting cultivation. A wide variety of crops is cultivated: in the wet season the staples maize, rice and millet are grown, while in the dry season people also rely on various fruits and vegetables. Communities that live adjacent to oxbow lakes or large rivers such as the Rufiji, Mbwemkuru, Kilombero or Ruaha practise fishing. Livestock keeping has never been Pachyderm No. 38 January–June 2005 Systematic recording of human–elephant conflict, Tanzania Mloka Mibuyusaba Mwaseni Nyaminywili Ndundunyikanza Msona Mtanza Kipo Ngorongo Kilimani 8°S R ufiji D istrict Tapika Namakono Kinjeketile Ngarambe Kipugira Kindunda Mtumbei Namatewa Kandawale Selous Game Reserve Njinjo Kikole Ruhatwe Miguruwe Mtepera 9°S De mocra tic R e public of C Chimbuko Barikiwa Kikulyungu Mkutano Uganda Kenya Soma lia K ilwa D istrict Rwanda Burundi Tanzania ongo i Ma la w Liwa le D istrict Ndunyungu Zambia Zimbabwe Mozambique Kinondoni 0 25 50 100 km Wailesi 10°S 38°E 39°E Figure 1. Human–elephant conflict study area between the Selous Game Reserve and the Indian Ocean. successfully carried out in the area because of the high prevalence of trypanosomiasis. Ethical taboos and religion have historically had an influence on controlling the consumptive use of some wildlife including elephants. Pachyderm No. 38 January–June 2005 Methods A standardized HEC data-collection protocol was followed to collect baseline data (Hoare 1999a). This protocol was developed and is endorsed by AfESG 31 Malima et al. and has been successfully used in several other HEC situations in Africa (Hoare 1999c; Parker and Osborn 2001; Sitati et al. 2003). Its aim is to quantify the distribution, frequency and severity of HEC incidents over large areas, do so reasonably cheaply, and achieve data of quality sufficient for local management of the problem. In this study, nine local enumerators were recruited and trained to capture primary data from conflict zones around the remote villages. Their training was carried out by the supervisor (C. Malima) in accordance with an AfESG-endorsed training package (Hoare 1999b) that is closely associated with the data protocol. When elephant damage occurs, the local HEC enumerator is informed through existing local communication networks. To maximize acceptance of the scheme among local communities, enumerators were assigned to collect HEC data only from the ward of their origin. The enumerator visits the incident site as soon as possible after the occurrence, travelling on foot or by bicycle. Enumerators discuss particulars of the incident with the affected people and quantify the property damage using standardized procedures. To avoid the problems associated with the use of paper maps, each enumerator was trained to use a GPS unit to record incident locations accurately. Apart from necessary particulars of date, location, farmers’ names and so on, the primary quantitative data captured from each incident are what crops the elephants damaged, how badly they damaged them, and if possible the group composition (age and sex) of problem elephants involved. The scheme began in July 2003 and the data presented here are those collected for one year, up to June 2004. Data collection in both wet and dry seasons monitored elephant damage to both seasonal and perennial crops. The seriousness of each crop-damage incident was subsequently further quantified by the supervisor, using a simple secondary data analysis based on the age, quality and damage level to the crop (Hoare 1999a). Crop damage by elephants was assigned to three levels—low, medium and high—by combining scores for the age of the crop and the quality and extent of the damage. Age categories for crops were given 1, 2 and 3 points for seedling, intermediate and mature growth stages. Quality categories of crops were given 1, 2 and 3 points for poor, medium and good. Damage was assigned to six categories (1–6 points) based 32 on percentage of crop-growing area damaged as assessed by the enumerator (≤5%; 6–10%; 11–20%; 21– 50%; 51–80%; >80%). The supervisor regularly checked a sample of incidents in situ, to minimize assessment bias by enumerators. The higher the score of combined points (age + quality + damage) for an incident, the more serious the damage suffered. All incident scores were then assigned to low (≤5 points), medium (6–8 points) or high (≥ 9 points) damage classes. Other categories of serious incident like human death or injury, damage to food stores or water sources and retaliatory killing of problem elephants were also recorded. The annual summary of many different conflict incidents gives a picture of the distribution, frequency and severity of the HEC problem in each ward and district. To quantify the proportion of farms affected by elephants, ideally all farms at risk should be mapped. In the prevailing conditions, however, this was not possible for a number of reasons: the very large number of individually owned plots; the extensive practice of shifting cultivation, and the limits of the workforce employed. Instead, locations (GPS waypoints) on the periphery of arable farming areas were recorded every 50 m and the total cultivated area used by village (including some fallow) was calculated by a computer program (ArcGIS 9.0 Desktop, ERSI 2004, Redlands, CA, USA). This was done in 26 of the villages. Results Types of incident In the 12-month period, 16 types of food crops were raided by elephants in 973 separate incidents (table 1). Farms commonly have mixed crops and during a raid elephants frequently damaged more than one crop type. Other conflict types identified (no. of cases in brackets) included: • people killed (2) or injured (1) by elephants • elephant damage to water sources (17) • elephants shot dead (25) by both wildlife officers and villagers • interference with people’s daily travel schedules such as obstructing children from attending school or restricting farmers moving to and from their fields (4) Pachyderm No. 38 January–June 2005 Systematic recording of human–elephant conflict, Tanzania Table 1. Elephant crop damage incidents in agricultural fields, July 2003–June 2004 Dry season Maize Millet Cashew nut Mixed crops Banana Cassava Rice Mango trees Peas Sugar cane Coconut trees Vegetables Sweet potato Orange trees Pawpaw Simsim Totals Wet season Rank Rufiji Kilwa Liwale Rufiji Kilwa Liwale Totals 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 20 5 21 34 46 0 — 7 3 9 — 9 0 0 2 — 4 0 24 23 30 4 — 45 23 16 18 0 6 5 2 — 0 2 89 15 2 70 — 1 5 3 — 1 1 0 0 — 122 36 — 17 10 — 44 — — 2 — 2 — — — 1 4 45 3 3 1 — 5 — 3 — 2 — — 1 — 2 25 51 — 26 1 8 12 — 2 — — — — — — — 175 139 137 118 90 82 61 53 36 30 20 12 7 6 4 3 156 200 189 234 69 125 973 The discrepancy in totals with tables 2 and 3 is due to multiple farms being damaged in some raids, non-assessment of some incidents, and counting of incidents other than crop raiding. Destruction of food stores by elephants, sometimes commonly encountered in other HEC zones in Africa, was not recorded during the reporting period. Seasonality of crop destruction by elephants Elephants seldom raided fields with crops at seedling stage. Raiding intensified in regularity and severity towards an annual peak as the harvest period approached for rainfed crops (June and July). Wetseason crop raiding damaged more annual crops like maize and rice, especially in the wetter Rufiji District. Faster-growing, more drought-resistant crops such as millet were damaged in the drier farming conditions of Kilwa and Liwale Districts. In the Rufiji Valley basin and to a lesser extent in Kilwa District, farmers carry out valley-bottom farming, growing mixed crops composed of maize, vegetables and peas on isolated farms. These areas were prone to dry-season elephant raids. In the dry season, farms with fruiting trees—cashew nut, mango, coconut and banana—were vulnerable to elephant invasion. Elephants debarked trees, broke branches to eat leaves or shook trees to dislodge fruit (especially cashew nut, mango and coconut). Perennial cassava, grown mostly in Liwale District, suffered dry-season damage. Pachyderm No. 38 January–June 2005 Level of elephant damage The year’s data were summarized by frequency (tables 1, 2) and severity (fig. 2) of incidents. Overall elephants caused relatively few cases of heavy losses in the three districts (5.8% wet season and 23.3% dry season). These high-level damage incidents often occurred in small and isolated agricultural fields, especially when they were raided by large groups of elephants. Most of the damage was in the mediumloss category (64.7% wet season and 56.3% dry season). Low-loss cases ranked intermediate (29.5% wet season and 20.4% dry season). Raiding group composition Raiding group size usually ranged from 3 to 10 but occasionally bigger groups of up to 40 animals were involved. Elephant mixed herds (bulls, cows and calves together) were the group type responsible for most of the crop raiding (table 3). Male groups (usually two to three animals) and lone bulls also caused considerable amounts of damage. In very few cases were cow–calf groups involved. These results should be interpreted with some caution, however, as data on the sexing of nocturnally active (and therefore mostly unseen) groups of raiders from their footprints are subject to error. 33 Malima et al. 300 Rufiji 242 Number of incidents 250 Kilwa Liwale 200 193 115 150 110 100 95 68 50 42 38 27 0 Low Medium High Elephant damage assessment Figure 2. Elephant crop damage levels assessed in the three districts, July 2003–June 2004. Variation between areas affected by elephants Elephant raids occurred in all nine wards and all but one out of 37 village areas (table 2). The worst crop damage was concentrated in Kikulyungu, Mloka, Chimbuko, Ngarambe and Miguruwe villages. These villages either border the SGR or are located in densely vegetated riverine habitats where elephants can easily take refuge. In Ngorongo Ward of Rufiji the relatively high level of elephant raiding in Kipo, Kipugira, Nyamnywili and Ndundunyikanza was possibly related to abandonment of fields following 32 human deaths caused by lions between October 2002 and April 2004—probably the most intense human–carnivore conflict in Africa (G. Packer, pers. comm. 2005). Wetand dry-season raids in Kikole village showed little difference. This village does not border the SGR but areas nearby harbour resident elephants year round. The low level of raiding in Kandawale Ward villages, which are also near the SGR, appears to coincide with elephant avoidance of the area as yet unexplained. Variation in the severity of elephant crop raiding can be judged by ranking villages. The simplest index is the number of incidents (table 2). But as many 34 incidents may not be serious, a more meaningful ranking for management priority is the number of raids in the high (or high and medium) damage categories. No meaningful relationship existed between size of area cultivated around villages (range 0.7–35 km2) and the number of elephant raids therein (range 0– 137) (table 2; fig. 3, R2 = 0.2112). Therefore a better ranking index to compare raiding intensity between villages could be: the size of the village area cultivated / number of raids (that is, raids per km2 of cultivation per year (Hoare 1999a). The range for raiding intensity in villages was also very wide at 0–25 incidents/km2 per year (table 2). In all villages combined, the overall raid intensity (1239 elephant ‘problem incidents’ in about 300 km2) was 4.1 incidents per km2 per year. Discussion The data presented here are from the first year of a study that is scheduled for three years, to capture between-year variation of this conflict. But even this first year’s results have successfully tested the principle and logistics of collecting primary data through an independent third party (the village-level enumerator), rather than from affected people’s verbal accounts Pachyderm No. 38 January–June 2005 Systematic recording of human–elephant conflict, Tanzania Table 2. Incident totals and ranking of elephant raids per village, July 2003–June 2004 District and ward Rufiji Ngorongo Mwaseni Utete Kilwa Kandawale Kikole Zinga-Miguruwe Liwale Barikiwa Kikulyungu Mpigamiti Totals: 9 wards Village Dry-season incidents Wet-season incidents Total incidents Ranking by no. incidentsa Ranking indexb Ngorongo Nyaminywili Kilimani 1 Kipo Kipugira Ndundunyikanza Kilimani 2 Mloka Mwaseni Mibuyusaba Mtanza Msona Ngarambe Tapika Nyamakono 25 23 23 0 0 0 0 73 8 21 11 6 31 11 0 21 14 24 9 4 8 24 62 21 13 7 16 49 0 1 46 37 47 9 4 8 24 135 29 34 18 22 80 11 1 10 12 9 24 28 22 16 2 14 13 18 17 4 22 29 8.2 4.9 4.9 2.6 — 5.0 22.4 12.7 5.3 25.2 — 3.8 21.4 8.3 1.4 Kandawale Kindunda Kinjeketile Mtumbei Namatewa Kikole Mbunga Nanyati Ruhatwe Migelgele Miguruwe Mtepera Naking’ombe Njinjo Zinga-Kibaoni 0 2 0 4 0 42 7 15 3 0 69 37 28 39 4 0 7 7 3 4 12 5 10 5 1 5 16 19 1 7 0 9 7 7 4 54 12 25 8 1 74 53 47 40 11 30 24 26 26 28 6 19 15 25 29 5 7 9 11 22 0 — 1.5 — — 1.6 — — 1.1 — 5.6 4.3 — 4.8 — Barikiwa Chimbuko Ndunyungu Kikulyungu Mkutano Kinondoni Wailesi 26 75 8 56 5 10 28 27 36 2 81 8 4 17 52 111 10 137 13 14 45 8 3 23 1 21 20 11 2.0 — 1.0 6.8 1.4 0.6 2.0 37 villages 692 547 1239 The discrepancy in totals with tables 1 and 2 is due to multiple farms being damaged in some raids, non-assessment of some incidents, and counting incidents other than crop raiding. a no. of raids b no. of raids per km2 of cultivated area — area not measured; therefore not ranked Pachyderm No. 38 January–June 2005 35 Malima et al. damage can be devastating to an individual farmer but its actual impact up the spatial scale through village and ward decreases progressively, until at disGroup type Wet season Dry season trict level its material impact is minor. Further no. % no. % assessments involving the economic value of losses (Tchamba 1996) at these scales, the seasonal selecMixed herd 271 56 435 63 tion of crops by elephants, and hypotheses tests of Lone bulls 108 22 94 13 causative factors in HEC will be done once the data Bull group 84 18 150 22 Cow–calf 21 4 12 2 of three years are available. Totals 484 691 Eastern Selous showed differences when compared with other areas where HEC has been systemThe discrepancy in totals with tables 1 and 2 is due to atically recorded (Parker and Osborn 2001; Sitati et multiple farms being damaged in some raids, nonal. 2003). Here the highest number of incidents was assessment of some incidents, and counting incidents other than crop raiding. caused by mixed herds as opposed to the more usual bull-only groups in fairly similar southern African woodland ecosys160 tems (Hoare 1999c, 2001b; Osborn 2003). Also, the medium category 120 of elephant damage was highest, R2 = 0.2112 rather than the low category—usu80 ally found most numerous elsewhere (Hoare 1999c; Parker and 40 Osborn 2001). High-category damage incidents were relatively few, 0 40 which agrees with findings else30 10 20 where (Hoare 1999c; Parker and 2 Area of cultivation in village (km ) Osborn 2001; Sitati et al. 2003). Figure 3. Elephant raids in the different size cultivated areas around The number of elephant raids 26 villages, July 2003–June 2004. per village showed a very weak positive relationship to the area of cultivation around villages, and the or from employees of a national wildlife authority. area-based index of raiding intensity varied widely. With both the latter, data are often inconsistent or bi- One such significantly positive relationship has been ased (Hoare 2001a). shown in very different land use in Kenya (Sitati et al. This study operated over probably the largest area 2003). But generally, quantitative associations with ever covered by a wildlife conflict reporting scheme of raid intensity are hard to find in the study of HEC, its kind (30,000 km2). Villages are widely spaced and suggesting alternative hypotheses that other spatial the proportion very small of the area actually occupied factors like habitat or crop type (Parker and Osborn and cultivated (1%)—and therefore in which HEC ac- 2001), farmers’ defences (Osborn and Parker 2002), tually occurred. Situations where small, scattered pock- and individual elephant behaviour (Hoare 1999c; ets of human habitation exist in a large matrix of natural 2001b) are more likely determinants of conflict levels. habitat containing elephants are fast disappearing in With only one year’s data, specific factors sigAfrica (Hoare and du Toit 1999), so this conflict assess- nificant in the damage pattern in the study villages ment gives a possible first-time indication of the HEC are speculative but may include the isolation of many scenario in historic times, for which only anecdotal ac- of the agricultural fields, the common practice of shiftcounts exist. ing cultivation, the severe damage to fruit trees and Despite what appear to be high raiding figures, vegetable plots, cultivation in the riverine habitats that actual economic losses were probably quite small and, elephants favour, and elephants moving in relatively in keeping with the pattern of this problem elsewhere large groups, which in turn may be due to human harin Africa, showed a gradient of seriousness. Elephant assment. No. of elephant raids Table 3. Elephant group type damaging crops, July 2003–June 2004 36 Pachyderm No. 38 January–June 2005 Systematic recording of human–elephant conflict, Tanzania A valuable finding was the relatively large number of elephants (25) that were killed in the conflict zone during the year. There is suspicion that HEC is sometimes being used as a pretext for illegal killings as some elephants were hunted by people from outside the area and meat was taken away to other parts of the country. Without the activities of an organized scheme systematically gathering data, it is doubtful that these killings would have been fully recorded in this large and remote area. This has important implications for other elephant ranges, especially sites involved in the CITES MIKE programme (www.citesmike.org), where collecting information exclusively on illegal activity may prove difficult; any activity that is perceived as helping people in conflict with wildlife has few opponents. The recording scheme proved cost efficient: it cost the equivalent of a little over USD 1000 to employ one enumerator for the year (USD 9180 to employ the nine enumerators for one year). Additional project costs were a small workshop to train enumerators and vehicle transport for regular field visits by the supervisor. The latter was outside the project budget but was met by the World Wide Fund for Nature International (WWF-International), who employed the supervisor, also responsible for other conservation work in the SGR. Most recording schemes using this model in smaller conflict zones can be run with six enumerators or fewer, and may cost less in other countries. The total area each enumerator can cover is dependent on terrain and land use, but in the past in the more common subsistence farming systems with scattered agriculture, 150–200 km2 per enumerator has been achieved (Hoare 1999c; Parker and Osborn 2001; Sitati et al. 2003). As the management of problem elephants is increasingly becoming de facto the responsibility of communities affected by them, it is especially important to quantify a conflict situation as much as possible, so as to be able to place it in a local conservation context (Hoare 2001a). The common research practice of using attitudinal questionnaire surveys (Kaltenborn et al. 2003; Holmern et al. 2004) is insufficient to understand wildlife conflict situations, as a clear disjunction has been shown to exist between perceived and actual problems (Languy 1996; Gillingham and Lee 2003). The scheme described here quantifies actual conflict incidents rapidly, impartially, cheaply and sufficiently accurately in rural African situations to be useful for local-scale wild- Pachyderm No. 38 January–June 2005 life management and land-use planning. For effective mitigation measures to be planned in any HEC situation (Hoare 2001a), initial collection of baseline information similar to that recorded in this study should be attempted. Acknowledgements We would very much like to thank Dr PJ Stephenson of WWF-International for encouragement and for soliciting funds for this study. WWF-Tanzania Programme Office Country Representative Dr Hermann Mwageni and Conservation Director Mr Stephen Mariki are sincerely thanked for their advice and support. Mr Leo Niskanen is thanked for his competent administrative support through the AfESG. We also appreciate the contribution and advice given by Mr Charles Mdoe, Assistant Director, Development, of the Wildlife Division. Enumerators who collected the field data were Messrs Abdallah Nnungu (Mpigamiti Ward), Adinani Rajabu (Zinga-Miguruwe Ward), Idd Ngajaja (Mkutano Ward), Juma Mpule (Barikiwa Ward), Mussa Maimbanya (Kikole Ward), Rajabu Lipanjanga (Kandawale Ward), Ramadhan Mneka (Ngorongo Ward), Said Kagoma (Utete Ward) and Saidina Malenda (Mwaseni Ward). We also thank the Wildlife Division district game officers, Messrs Eniyoye J. John (Rufiji District), Chande M. Ligibu (Kilwa District) and Said Kabanda (Liwale District) for working very closely with the enumerators. Three anonymous reviewers made useful comments on the manuscript and Mr Simon Mwansasu assisted with data for the map. The corresponding author thanks his employer, the Messerli Foundation of Switzerland, for the use of their facilities while writing the paper. References Blanc JJ, Thouless CR, Hart JA, Dublin HT, Douglas-Hamilton I, Craig CG, Barnes RFW. 2003. African elephant status report 2002: an update from the African Elephant Database. IUCN/SSC African Elephant Specialist Group, Gland, Switzerland. Dublin HT, Hoare RE. 2004. Searching for solutions: an integrated approach to understanding and mitigating human–elephant conflict in Africa. Human Dimensions of Wildlife 9:271–278. Gillingham S, Lee PC. 2003. People and protected areas: a study of local perceptions of wildlife crop damage con- 37 Malima et al. flict in an area bordering the Selous Game Reserve, Tanzania. Oryx 37(3):316–325. Hoare RE. 1999a. Data collection and analysis protocol for human–elephant conflict situations in Africa. IUCN African Elephant Specialist Group, Nairobi, Kenya, and www. iucn.org/afesg. 37 p. Also in French. Hoare RE. 1999b. Training package for enumerators of elephant damage. IUCN African Elephant Specialist Group, Nairobi, Kenya, and www. iucn.org/afesg. 17 p. Also in French. Hoare RE. 1999c. Determinants of human–elephant conflict in a land-use mosaic. Journal of Applied Ecology 36:689–700. Hoare RE. 2001a. A decision-support system for managing human–elephant conflict situations in Africa. African Elephant Specialist Group, Nairobi, Kenya, and www. iucn.org/afesg. 104 p. Also in French and Portuguese. Hoare RE. 2001b. Management implications of new research on problem elephants. Pachyderm 30:44–48. Hoare RE, du Toit JT. 1999. Coexistence between people and elephants in African savannas. Conservation Biology 13(3):633–639. Holmern T, Johannesen AB, Mbaruka J, Mkama S, Muya J, Roskaft E. 2004. Human–wildlife conflicts and hunting in the western Serengeti, Tanzania. Norwegian Institute for Nature Research (NINA) Project Report No. 26. Tungasletta 2, NO-7485 Trondheim, Norway, and www.nina.no Kaltenborn BP, Nyahongo JW, Mayengo M. 2003. People and wildlife interactions around Serengeti National Park, Tanzania. Norwegian Institute for Nature Research (NINA) Project Report No. 22. Tungasletta 2, NO-7485 Trondheim, Norway, and www.nina.no 38 Languy M. 1996. Suivi et attenuation de l’impact des éléphants et autres mammifères sauvage sur l’agriculture au Gabon. Rapport final, WWF Programme pour le Gabon. Matzke G. 1976. The development of the Selous Game Reserve. Tanzania Notes and Records No. 79 and 80. Mpanduji DG, Hofer H, Hildebrandt TB, Goeritz F, East ML. 2002. Movement of elephants in the Selous–Niassa wildlife corridor, southern Tanzania. Pachyderm 33:18–31. Nicholson BD. 1969. The Selous Game Reserve. Unpublished mimeograph copy of a speech to 2nd annual meeting of Game Conservation International, San Antonio, Texas, USA. Osborn FV, Parker GE. 2002. Community-based methods to reduce crop losses to elephants: experiments in the communal lands of Zimbabwe. Pachyderm 33:32–38. Osborn FV. 2003. Seasonal influence of rainfall and crops on home-range expansion by bull elephants. Pachyderm 35:53–59. Parker GE, Osborn FV. 2001. Dual-season crop damage by elephants in the eastern Zambezi Valley, Zimbabwe. Pachyderm 30:49–56. Sitati NW, Walpole J, Smith RJ, Leader-Williams N. 2003. Predicting spatial aspects of human–elephant conflict. Journal of Applied Ecology 40:667– 677. Tanzania Census. 2002. Population and housing census general report. Central Census Office, National Bureau of Statistics, Dar es Salaam. Tchamba MN. 1996. History and present status of the human–elephant conflict in the Waza–Logone region, Cameroon. Biological Conservation 75:34–41. Pachyderm No. 38 January–June 2005 Elephant crop damage in the Red Volta Valley, Ghana Elephant crop damage in the Red Volta Valley, north-eastern Ghana Patrick Adjewodah,1* Paul Beier,2 Moses K. Sam3 and John J. Mason1 1 Nature Conservation Research Centre, PO Box KN 925 Kaneshie, Accra, Ghana; email: [email protected] and [email protected]; *corresponding author 2 School of Forestry, Northern Arizona University, Flagstaff, AZ 86011-5018, USA; email: [email protected] 3 Resource Management Support Centre, Forestry Commission, PO Box 1457, Kumasi, Ghana; email: [email protected] Abstract We monitored the crop-raiding behaviour of elephants that seasonally migrate into the Red Volta Valley as part of a project to mitigate conflict. During 1999–2003 we organized farmers in the project area into associations, recorded the size of each raided farm, crops affected, area of farm damaged, date of damage and farm location. The percentage of registered farmers affected by elephant crop raiding ranged from 2 to 3% per year. Mean area damaged per raided farm averaged 0.98 ha in 2002. Damage varied considerably among affected farms with < 1% to 100% of the cultivated area destroyed. Most raiding incidents occurred between August and November, before and during crop harvest, which is also the period elephants migrate from Burkina Faso into the study area. Seasonal migration produced a strong geographic pattern of sequential raiding in chiefdoms bordering the Red Volta forest reserves that serve as the main elephant corridor. Chiefdoms farthest from the Ghana–Burkina Faso border were the last to experience crop raiding each year. Within affected chiefdoms, proximity to the forest boundary and the Red Volta River probably increases the risk of crop raiding. Human– elephant conflict can be reduced by planting crops that mature early to advance harvest time to occur before elephants arrive, and by locating farms away from forest reserves. Additional key words: farm monitoring, elephant corridor Résumé Nous avons surveillé le comportement des éléphants qui ravagent les cultures et qui migrent de façon saisonnière dans la Vallée de la Volta rouge, dans le cadre d’un projet destiné à tempérer les conflits. De 1999 à 2003, nous avons rassemblé les fermiers de la zone du projet en associations, nous avons enregistré la taille de chaque ferme attaquée, les cultures ravagés, la surface de ferme touché, la date des dommages et l’emplacement de la ferme. Le pourcentage des fermiers enregistrés affectés par les dégâts d’éléphants allait de 2 à 3% par an. La surface moyenne touchée par attaque d’éléphant était de 0,98 ha en 2002. Les dégâts variaient considérablement selon les fermes, allant de < 1 à 100 % de la surface cultivée. La plupart des incidents se passaient entre août et novembre, avant et pendant les récoltes, ce qui correspond au moment où les éléphants migrent du Burkina Faso vers la région étudiée. La migration saisonnière a produit un pattern géographique très marqué de raids séquentiels dans les chefferies qui bordent les réserves forestières de la Volta Rouge qui servent de corridor principal pour les éléphants. Les chefferies les plus éloignées de la frontière Ghana–Burkina Faso sont chaque année les dernières à subir les raids des éléphants. A l’intérieur des chefferies, la proximité de la limite forestière et du fleuve augmente probablement les risques de dommages. On pourrait réduire les conflits hommes– éléphants en plantant des semences qui mûrissent tôt, pour avancer le moment des récoltes avant l’arrivée des éléphants, et en installant les fermes plus loin des réserves forestières. Mots clé supplémentaires : surveillance pour les fermes, corridor d’éléphants Pachyderm No. 38 January–June 2005 39 Adjewodah et al. Introduction Loss of habitat, illegal trade in ivory, and human intolerance to crop raiding are the major obstacles to conserving both African and Asian elephants (Sukumar 2003). Crop raiding occurs throughout elephant ranges and probably began with the advent of agriculture 10,000 years ago. It has intensified as agriculture has spread throughout the elephants’ range. The scientific study of crop damage by elephants began only in the 1970s in Asia and in the 1980s in Africa. Since then, a number of deterrent methods have been developed, including disturbance shooting and electric fencing (Kangwana 1995; Thouless and Sakwa 1995). In the 1970s, electric fencing was considered the best way to stop crop-raiding elephants, but not all fences have worked (Thouless and Sakwa 1995). The high cost of constructing and maintaining electric fences is unrealistic for many elephant-conflict sites in Africa. Disturbance shooting has been widely applied since the colonial era despite suggestions that it also is only minimally successful in mitigating crop raiding (Ayigsi 1997; Osborn and Parker 2002). In 1970 the Ghana Wildlife Division (a unit of Ghana’s Forestry Commission), under pressure from complaints of crop raiding in the Red Volta Valley, undertook a severe culling operation with the aim of eliminating the elephant problem from the area (B. Jamieson, consultant, 2000, pers. comm.). The cull reportedly continued until no evidence of elephants remained in the river valley. But by the mid-1990s elephants were back and the Wildlife Division used disturbance shooting to scare them away from farms. However, the Wildlife Division did not have permanent presence in the valley and instead dispatched deterrent teams from distant locations in response to complaints from farmers and local politicians. These ad hoc interventions were limited by budgetary constraints and had little impact on the conflict (Ayigsi 1997; NCRC 1999). In 1999 a community-based approach that empowered local people to implement low-cost deterrent measures was initiated and has since been increasingly effective in mitigating the problem (Adjewodah et al. 2003). Causes of crop raiding Several factors affect the number and location of human–elephant conflicts. In Africa seasonal movements of elephants bring them into contact with 40 farmlands, which have encroached and fragmented on their traditional range (Hoare 1999). In India seasonal elephant movement, competition for water, reduction and degradation of natural habitat, and the higher nutritive value of cultivated crops as compared with uncultivated food are associated with increased crop raiding (Sukumar 1990). In Zimbabwe rainfall and plant moisture may influence the movement of elephants into communal land from a protected area (Osborn 2003). Insufficient habitat in protected areas and modification of the landscape by humans contribute to elephant crop raiding in the Upper Guinean forest zone of West Africa (Barnes 2002). In this same landscape Sam et al. (2003) has suggested that elephant migratory movements affect crop raiding. Sam et al. (1998) found that a growing human population and the need for new farmland has increased human–elephant conflict in the Red Volta Valley. Others (Lowry and Donahue 1994; Okoumassou et al. 1998) attributed a surge in the incidence of elephant crop raiding in the Red Volta Valley in the early to mid-1990s to the displacement of elephants in northern Togo during political instability in that country. History of crop raiding In the mid-1990s, crop raiding by elephants was an important problem for farmers in the Red Volta Valley (NCRC 1999). Okoumassou et al. (1998) reported farmers were intolerant to the risk of losing crops to elephants and suggested that local people’s hostility was the most critical short-term threat facing elephants there. Farmers’ intolerance to crop damage encouraged elephant hunters employed to protect communal farm enclaves to poach (NCRC 1999, 2000; Adjewodah et al. 2003). The Red Volta Valley Conservation Project, which was initiated in 1999 to mitigate the conflict that ensued, involved the collaboration of the Nature Conservation Research Centre (NCRC—a Ghanaian conservation NGO), the communities and traditional authorities in the project area, the Wildlife Division, and the Bolgatanga and Bawku West district assemblies among others. In this paper, we report on patterns of crop damage during 1999–2003. Our objectives are to describe the patterns of crop raiding by an elephant herd that is only seasonally present in the region, and determine whether crop damage is related to geographic location, stage of crop maturity and type of crops. Pachyderm No. 38 January–June 2005 Elephant crop damage in the Red Volta Valley, Ghana Research area The Red Volta Valley comprises the Red Volta East, Red Volta West, Gambaga Scarp East, Gambaga Scarp West, Morago East and Morago West Forest Reserves and adjacent woodlands and fallow lands off the reserve (latitude 10°3’2 to 11°00’2 N, longitude 0°45’2 to 0°00’2 W) in the Bolgatanga, Bawku West and Bongo Districts of the Upper East Region (fig. 1). The area experiences two climatic seasons: dry and wet. The wet season extends from May to November, and the dry season from December to April. The mean annual rainfall is about 900 mm with an annual peak between July and September. The dry period is characterized by desiccating north-east winds known as harmattan, which bring dust and haze from the Sahara Desert, and by bushfires between December and February affecting over 90% of the area annually. The vegetation is fire pro-climax and is locally influenced by human activities such as farming, charcoal burning, small-scale mining, cattle grazing, and harvesting of firewood and building poles (Wheelan 1950). BURKINA FASO R Widnaba E ta Vol Red Bongo The Red Volta, White Volta and Morago Rivers and their tributaries drain the area. Both the Red Volta (called River Nazinon in Burkina Faso) and the White Volta flow south from Burkina Faso. The Morago takes its source from northern Togo near Fosse aux Lions National Park as River Koulagouna, and flows into the White Volta River in Ghana (fig. 1). The gallery of forest reserves along these rivers and the adjoining community lands are the major areas with natural vegetation, which is predominantly Guinean savanna woodland (Taylor 1952) that constitutes a network of habitat linking the Red Volta Valley to elephant ranges in northern Togo and southern Burkina Faso. The Red Volta area has been further discussed elsewhere (Sam et al. 1998). Okoumassou et al. (1998) reported the movement of elephants between Ghana and Togo along the Morago–Koulagouna Rivers. Elephants were seen most frequently in the Red Volta Valley during the wet season, especially as harvest time approached. Okoumassou et al. (1998) noted that elephants seem to move northwards from the Red Volta into Burkina Faso for the dry season, and southwards again in the Zebiila Natinga D bo Tam V O Tilli Garu Natinga L Nangodi T a olt eV hit W A W T E S Sakote Bulpielisa ta Vol ed R T F O R E S Datoko R V E S E R E Tongo Kusanaba RED VOLTA EAST FOREST RESERVE ta Vol ite Wh a olt dV Re e hit W lta Vo GAMBAGA SCARP WEST FOREST RESERVE MORAGO EAST FOREST RESERVE MORAGO WEST FOREST RESERVE Zongoiri Biungu/Degari RED VOLTA WEST FOREST RESERVE lta White Vo Shiego Bugwia ta Vol ite Wh GAMBAGA SCARP EAST FOREST RESERVE Gambaga Nakpanduri Bimbago Sakogu Nalerigo o rag Mo TOGO er Riv LEGEND road river international boundary town/village forest raided farms Figure 1. Map of villages and chiefdoms showing their position relative to the Ghana border and to the Red Volta River and forest reserves. Pachyderm No. 38 January–June 2005 41 Adjewodah et al. early wet season. In the early 1990s elephants were common and resident in northern Togo, but political unrest during 1990–1992 in that country encouraged encroachment of protected areas and displaced elephants into neighbouring Benin, Ghana and Burkina Faso (Lowry and Donahue 1994; Okoumassou et al. 1998). The more than 130 elephants inhabiting the Togo end of the corridor are believed to have disappeared (Okoumassou et al. 1998), and the seasonal migration of elephants across the border between Red Volta Valley and northern Togo appears to have halted (Adjewodah 2004). Farming practices Farming in the study area involved seasonal rain-fed subsistence agriculture. Farming activities started with land preparation beginning in the late dry season (March–April), followed by planting in May and June. Farm sizes in four chiefdoms in 2003 ranged from 0.1 ha to 7.3 ha (mean 0.7 ha, n = 104). Farm size as estimated and reported by farmers in seven chiefdoms averaged 1.3 ha (n = 814). Crops grown, in order of decreasing area cultivated, include millet, maize, groundnut, bean, guinea corn, rice and yam (table 1). In 2003 monocultures of millet comprised 33%, maize 30% and groundnut 12% (n = 104 farms) of the area of crops grown. A variety of early-maturing millet usually grown in June (in the courtyard of the farmer) takes about two months to mature. A second variety of millet, known as late millet, is sown after June on bush farms (farms closer to the forest reserves than to settle- ments). This variety is harvested between September and November. Methods To survey crop damage in farms throughout the project area we organized farmers into associations called farm-monitoring groups, within traditional authority constituencies or chiefdoms. An average of 8 (range 7–9) chiefdoms participated each year; here we report on 7 chiefdoms for which consistent records were available. Each constituency or chiefdom consisted of 4 to 12 villages, which shared a common farm enclave and a local chief. The communal farm enclaves were adjacent to the forest reserves and consisted of islands of cultivation within fallows. We did not register farmers cultivating enclaves where crop raiding was not an issue or farms inside the reserves. We registered 996 farmers in 1999, the first year of the project. The number of participating farmers increased to 1500 in 2000 following a perceived drastic reduction in crop-raiding incidents (NCRC 2000; Adjewodah et al. 2003). In 2003, 1030 farmers were registered (table 1). Although we did not enumerate the entire farming population, we believe over 80% of farmers cultivating farmlands with raiding history participated in the project each year. In some areas, the improved confidence of farmers resulted in the return to abandoned farms, some of which were on the edge of the forest reserves. Table 1. Area of farms (in ha) reported by registered farmers in 2003 (total for all crops exceeds total area because some crops were grown together) Chiefdom Biungu Kusanaba Nangodi Sakote Tilli Widnaba Zongoiri Total Farmers Area registered cultivateda Millet Maize Groundnut Crop Bean Guinea corn Rice Yam 74 160 142 246 94 127 187 498 114 –b 158 56 91 607 498 0 –b 1 1 74 468 498 70 –b 14 54 4 125 119 8 –b 106 1 22 27 219 0 –b 5 0 7 0 20 105 –b 17 14 8 0 56 0 –b 20 0 1 0 62 0 –b 0 0 0 0 1030 1524 1042 765 283 231 164 77 62 a Area cultivated by registered farmers only. The chiefdom’s farm enclave was about 3600 ha in each of Biungu, Kusanaba, and Zongoiri, about 1600 ha in each of Nangodi and Sakote, 2600 ha in Tilli, and 2000 ha in Widnaba. We did not encourage farmers to register where crop raiding was not an issue or if they were inside the reserves. b Data not recorded 42 Pachyderm No. 38 January–June 2005 Elephant crop damage in the Red Volta Valley, Ghana We held a series of meetings to introduce farmers to the monitoring programme and a companion programme to reduce crop damage. Registration was voluntary. Each of the 33 groups participating in the programme in 1999 elected two representatives: a leader and an enumerator. Members of the farm groups and associations registered their farms with their respective enumerators during the planting seasons (May/June) of 1999 to 2003. We supervised these exercises and made sure the following farm variables as reported by the farmer were recorded: name of farmer, sex, general location of farm, size of farm, crops planted and date of sowing. The farm groups cooperated to implement a set of measures against crop raiding by elephants (Adjewodah et al. 2003). Group members shared information on elephant damage and movement among themselves and with the project team, and they formed cooperatives to enable them to harvest crops in a timely manner. The project team held periodic meetings with the groups to prepare them for the cropraiding season. The effectiveness of these measures was gauged by a monitoring exercise, which required the registered farmers to record the crop damage they reported. Affected farmers reported crop-raiding incidents to their leader or enumerator. The enumerator then visited the affected farm to assess and record the damage using a standardized format developed by the IUCN Human Elephant Working Group of the African Elephant Specialist Group (Hoare 1999). The enumerator estimated the area damaged by pacing it off; categorized crop stage as seedling, intermediate, or mature; and graded the quality of crops affected as poor, medium or good. This assessment was subjective because it relied on the judgement of the enumerator and opportunistic because it relied on farmers to report incidents. These weaknesses were, however, outweighed by affording large coverage of the elephant range, low cost and sustainability of approach. We collected data forms for analysis from the group leaders at the end of the crop-raiding season in November. Because records for 1999 were incomplete, 1999 data were not used in some analyses. We used an index of damage developed by Hoare (1999). The damage score is the sum of the age score of crop (1 = seedling, 2 = intermediate, 3 = mature), the quality score (1 = poor, 2 = medium, and 3 = good) and the damage category (1 = ≤5% of farm area damaged, 2 = 6–10%, 3 = 11–20%, 4 = 21–50%, 5 = 51– Pachyderm No. 38 January–June 2005 80%, and 6 = >80%). We interpreted damage scores as low (1–5), medium (6–8), or high (9–12). We recorded the geographical coordinates of some villages and raided farms with a hand-held geographical positioning system. We used a GIS to calculate the distance of settlements and raided farms from the nearest forest reserve boundary and from the Red Volta River. In 2003 only, we randomly selected and measured 20% of the registered farms in the two chiefdoms most affected (Kusanaba and Sakote) and the two least affected (Tilli and Widnaba). For each of the 104 farms, we recorded GPS coordinates, measured the cultivated area (by pacing) and recorded the crops raised. Results Elephant damage Each year less than 4% of registered farmers were affected and the mean damaged area per raided farm was less than 1.5 ha (table 2). Damage varied considerably among farms and ranged from less than 1% to 100%, with about 56% of raided farms experiencing damage to less than 21% of the farm and about 27% of raided farms experiencing damage to over 50% of the farm (fig. 2). During 1999–2002 elephants raided 111 registered farms affecting 2.3% of the farms registered during this period (table 2). The percentage of farmers affected ranged from 2 to 3% per year with little change among years until 2003, when elephants were absent in the Red Volta during harvest. The mean damage score of affected farms was low (table 2). Between 2000 and 2002 millet was the most affected crop followed by guinea corn, groundnut and rice (table 3). The affected crops were usually raided during mature stages when they were ready for harvest. Most raiding affected crops of medium to good quality with considerable variation among crops (table 3). Temporal and geographic pattern of crop raiding Crop raiding by elephants usually occurred from June to November. October was the peak crop-raiding month in each year (fig. 3) with about 72% of the raiding cases during 2000 to 2002 being recorded. The peak period 43 Adjewodah et al. Table 2. Number of farmers registered and affected by elephant crop raiding, and mean annual damage score for 1999–2003 Number affected Year Number of registered farmers Percentage of farmers affected 1999 2000 2001 2002 2003 996 1500 1500 859 1030 30 32 24 25 0 3 2 2 3 0 Average number of Mean damage score hectares damaged per per raided farm (on raided farm scale of 1–12) not available 1.29 0.53 0.98 – not available 1.39 1.31 1.25 – 40 Percentage of farms chiefdoms further south (fig. 1). In 2000 the first cropraiding incident was at 30 Kusanaba on 2 September, followed by raids in Sakote later in the same month. It was not until 2 October that Biungu, the chiefdom far20 thest from the Ghana– Burkina Faso border, experienced its first raid that year. In 2001 the first raided 10 chiefdoms were Nangodi and Sakote. Zongoiri, the chiefdom farthest from the border, experienced its first raid on 0 15 November, about two <1 11–20 31–40 51–60 81–90 months after the first reported 1–10 21–30 41–50 71–80 91–100 cases at Sakote and Nangodi. A similar pattern occurred in Percentage of farms damaged 2002. This pattern is consistFigure 2. Frequency distribution of percentage of farm damage for 87 farms ent with an annual movement raided during 2000–2002. from Burkina Faso into the Red Volta Valley (fig. 1). All crops the elephants raided were on bushfarms coincided with the time when farm crops were mature, ripe or ready for harvest. However, in 2002 raiding about 2–8 km from the nearest village. Compared with started as early as June (28%) and ended in November the main village in each chief-dom, farms damaged by elephants were relatively close to the Red Volta (4%). Crop raiding was localized, with elephants return- West and Red Volta East Reserves (fig. 1). Because ing to the same communities over the years. The Sakote only farmers historically affected by elephants were chiefdom recorded 45 of the 90 raids between 2000 likely to register, we could not further quantify proxand 2002; next were 18 raiding incidents in Kusanaba, imity to forest reserves as a risk factor. Overall, there 9 in Nangodi and 9 in Biungu (table 4). The annual was no correlation between the percentage of regisincidents per chiefdom were low near Ghana’s border tered farms in a chiefdom that was raided by elephants with Burkina Faso, where the elephants are assumed to and the distance from the main village in the chiefdom spend most of the year, peaked further south along the to the Red Volta River (r = –0.28, p > 0.50, n = 7). Red Volta at Sakote and Kusanaba, and decreased in For the two most heavily raided chiefdoms (Sakote 44 Pachyderm No. 38 January–June 2005 Elephant crop damage in the Red Volta Valley, Ghana Table 3. Composition of damaged area for 105 farms raided by elephants during 2000–2002 Crop Percentage of crop area damaged Millet Guinea corn Groundnut Rice Maize Bean 31 27 17 16 8 1 Percentage of crop in mature intermediate seedling stage stage stage 71 42 100 16 68 100 29 58 0 79 32 0 0 0 0 5 0 0 Percentage of crop of good medium poor quality quality quality 38 56 94 32 43 100 44 40 0 37 43 0 18 4 6 31 14 0 have been reported). Furthermore, most raided farms suffered < 20% crop loss. Apparently, crop damage in the 60 2000 Red Volta region of Ghana has 2001 remained low since 1999 (see 2002 below), and is no longer important in the regional economy as 40 a whole. However, farms are small, the cultivated field is the farmer’s investment for the 20 whole year, and the impact on individual farmers can be considerable. Even a small area raided is catastrophic to the 0 farmer and the farm family. June August September October November Farmers identify with the plight of their neighbours, and crop Month raiding affecting one family Figure 3. Percentage of annual crop-raiding incidents occurring in each may generate animosity towards month by year during 2000–2002. elephants in many. Thus continual efforts should be made to and Kusanaba), we calculated distances of a random maintain the current low level of raiding incidents sample of registered farms from forest reserves. For and harmony between farmers and elephants. DiverSakote, the mean distance of farms raided during sifying economic activities in communities along the 2000–2002 (0.58 km) was less than the mean dis- corridor could make a significant difference in how tance of registered farms from forest reserves (1.27 farmers perceive elephants. Unlike most human–elephant conflict zones in km). However, these distances did not differ at Kusanaba with 2.23 km for random farms and 2.33 Africa, crop raiding in the Red Volta is predictable because of the seasonal migration. Raiding incidents km for raided farms. are concentrated in a few months close to the time crops are harvested. Thus, simple measures such as Discussion planting crops that can be harvested early helped reIn most years about 2–3% of registered farms were duce crop raiding to the low levels reported here (2– raided by elephants. Because our programme attracted 3% of registered farmers) from higher levels in the only farmers at risk of elephant crop raiding, the over- early 1990s (Adjewodah et al. 2003). Although there all incidence of crop raiding in the chiefdoms is prob- were no reliable baseline data for assessing crop damably lower than this (although some raids may not age before the project intervention, NCRC (2000) Percentage 80 Pachyderm No. 38 January–June 2005 45 Adjewodah et al. Table 4. Frequency and extent of crop raiding in relation to distance of chiefdoms from the nearest forest reserve boundary and the Red Volta River Main village in chiefdom Biungu Kusanaba Nangodi Sakote Tilli Widnaba Zongoiri Total Distance from Distance to main village to Red Volta forest reserve (km) (km) 4.36 7.10a 1.28 1.58b 2.62c 2.16d 4.24 7.99 8.26 5.24 4.81 5.33 6.78 14.45 Number of incidents Mean area damaged per affected farm (ha) 2000 2001 2002 Total 2000 2001 2002 Total 8 2 0 28 0 0 0 0 12 2 5 1 0 6 1 4 7 12 2 0 0 9 18 9 45 3 0 6 0.09 1.04 0.00 0.16 0.00 0.00 0.00 0.00 0.09 0.15 0.02 0.05 0.00 0.21 0.00 0.01 0.75 0.11 0.02 0.00 0.00 0.09 1.14 0.90 0.29 0.07 0.00 0.21 38 26 26 90 1.29 0.52 0.89 2.70 a Compared with the main village in each chiefdom, most registered farms were located closer to the river and to forest reserves. For a random sample of 32 registered farms in this chiefdom in 2003, distances to forest reserve ranged from 0.63 to 3.83 km (mean 2.23). b For a random sample of 34 registered farms in this chiefdom in 2003, distances to forest reserve ranged from 0.49 to 1.88 km (mean 1.27). c For a random sample of 18 registered farms in this chiefdom in 2003, distances to forest reserve ranged from 1.23 to 5.48 km (mean 2.65). d For a random sample of 20 registered farms in this chiefdom in 2003, distances to forest reserve ranged from 0.47 to 2.16 km (mean 1.01). estimated from farmer interviews that about 400 ha of farm were destroyed by elephants over a two-year period (1995–1996) compared with a documented total of 18 ha during 2000–2002. However, we caution that crop-damage data collected through interviews are most likely an exaggeration and may not provide a reliable basis for comparison with our quantitative data. But as many as 133 farms were damaged in 1996 (Sam et al. 1997), compared with an average of 22.2 farms affected from 1999 to 2003 (table 2). On the other hand, it is possible the decline in crop raiding is a reflection of declining numbers of elephants migrating from Burkina Faso. The lack of incidents in 2003 is likely because elephants apparently that year did not enter Ghana when crops were at the preferred stage of growth (Adjewodah 2004). However, migratory elephants were recorded within 7 km of the Ghana border in 2003 (Sawadogo 2003) and NCRC staff have recorded their return in 2004. Location of chiefdoms or farm enclaves with respect to the Ghana–Burkina Faso border (on a north– south axis) and to the forest reserves (on an east–west axis) are among factors that determine timing and frequency of elephant crop raiding within chiefdoms. Each year, the annual events of crop damage start with elephant raids in chiefdoms close to the border in- 46 cluding Sakote, Tilli and Kusanaba. This pattern probably reflects the cross-border migration and movement of elephants from south-central Burkina Faso to the Red Volta Valley. However, Widnaba, the chiefdom nearest to the border, experienced no crop raiding from 2000 to 2002, indicating that distance from the border is not the only factor. Sam et al. (2002) found that an increasing number of cattle reduces the probability of finding elephants outside the Red Volta forest reserves. Elephants may also avoid areas with increased human presence and disturbance (Barnes et al. 1991; Mpanduji et al. 2002), high poaching or prevalence of physical barriers. However, we have no evidence that Widnaba differs from other communities with respect to these factors. Within chiefdoms, farms closest to forest reserves and the major rivers are at increased risk of crop damage by elephants. Our observation is consistent with results from elsewhere in Africa. Bell (1984) mentions animals using rivers as channels for crop damage. Naughton-Treves (1997) mentions heavier crop damage in fields at the edge of forests. Parker and Osborn (2001) note increased crop damage at the edge of protected areas and increased crop raiding along river systems in the dry season. Although we lacked the data (that is, locations of a large sample of farms) needed to quantify how risk varies with distance from Pachyderm No. 38 January–June 2005 Elephant crop damage in the Red Volta Valley, Ghana forest, the very fact that only farmers near forest reserves registered for the programme indicates the importance of such proximity. This almost certainly reflects elephant preference for these rivers and forests as a migratory corridor (Sam et al. 1997) and underscores the importance of maintaining natural vegetation in these areas against farm encroachment and unnaturally frequent bush fires. We recommend moving some farms away from the reserve boundary to further reduce elephant raiding in the Red Volta Valley. However, crop raiding will continue to remain an issue as long as elephants remain in the valley because they prefer cultivated grasses, such as finger millet, to wild grass, as they seek more nutritious foods in line with the optimum forage theory (Sukumar 1990, 1994). Again in line with the optimum forage theory, our results indicate elephant preference for mature and good-quality crops, which are likely to be more nutritious than immature crops. Bell (1984) and Parker and Osborn (2001) also found that elephants select mature crops over seedlings or immature crops. Thus it will remain important to organize farmers at risk into groups that can share knowledge and quickly haze elephants or harvest crops in response to approaching elephants. Conclusion The results from our survey also underscore the urgent need for measures to conserve the Red Volta Valley. This area holds a small and vulnerable population of elephants that seasonally migrate between Ghana and Burkina Faso (and to Togo in the recent past). The key to successful management of the valley and the elephants within it lies in a collaborative effort by stakeholders both in-country and across the borders. A collaborative community reserve approach that will make local communities decision-makers not onlookers, that will make them benefactors from elephants not victims of their activities, holds a promising future for elephants. We support efforts by IUCN/AfESG to facilitate international collaboration among Ghana, Burkina Faso and Togo to create the Nazinga–Kabore Tambi– Red Volta–Doung elephant corridor to formally link the Red Volta Valley with elephant ranges in southcentral Burkina Faso and northern Togo. In this regard, we have developed a concept paper for the Ghana component of this corridor, which will be validated with a broad group of stakeholders and then Pachyderm No. 38 January–June 2005 developed into an implementation plan. The Earthwatch Institute, a charitable international institution supporting field research, is also working with us on a longterm ecological research project in the Red Volta starting from 2005, which will advance Ghana’s effort to develop the elephant corridor. Acknowledgements The 2003 crop-raiding data collection was supported by the European Commission through the Small Grants Programme of the African Elephant Specialist Group (AfESG) of the Species Survival Commission (SSC), World Conservation Union (IUCN). The kind support of the above institutions is deeply appreciated. The Red Volta Conservation Project is also indebted to USAID and the Canada Fund for the seed funding of this project. We are grateful to Drs R. Sukumar and R.F.W. Barnes for providing insight and technical papers to Mr P. Adjewodah during project planning. We are grateful to Messrs G. Agbango, the project field assistant, and A.J. Murphy for playing their various roles. Finally we especially thank the project communities for their cooperation and hospitality. We hope that this paper will encourage ongoing efforts and draw new partners in supporting efforts towards coexistence of farmers and elephants in the Red Volta Valley. References Adjewodah P. 2004. Habitat status, population and distribution of the African savanna elephant (Loxodonta africana) in north-eastern Ghana. Unpublished report. IUCN/AfESG project SG0203, Nairobi. Adjewodah P, Mason JJ, Murphy AJ. 2003. Mitigating elephant crop raiding: the Red Volta Valley experience. Unpublished report. IUCN/AfESG, Nairobi, Kenya. Ayigsi J. 1997. Volta–Morago community elephant reserve project: summary of field report. Unpublished project report. Nature Conservation Research Centre, Accra. Barnes RFW. 2002. Treating crop-raiding elephants with aspirin. Pachyderm 33:96–99. Barnes RFW, Barnes KL, Alers MPT, Blom A. 1991. Man determines the distribution of elephants in the rainforest of northern Gabon. African Journal of Ecology 29:54–63. Bell RHV. 1984. The man–animal interface: an assessment of crop damage and wildlife control. In: Bell RHV and Mcshane-Caluzi, eds., Conservation and wildlife management in Africa. US Peace Corps seminar, Malawi. 47 Adjewodah et al. Hoare R. 1999. Data collection and analysis protocol for human–elephant conflict situation in Africa. A document prepared for the IUCN African Elephant Specialist Group’s Human–Elephant Conflict Taskforce. AfESG, Nairobi. Kangwana K. 1995. Human–elephant conflict: the challenge ahead. Pachyderm 19:11–14. Lowry A, Donahue TP. 1994. Parks, politics and pluralism: the demise of national parks in Togo. Society and Natural Resources 7:321–329. Mpanduji DG, Hofer H, Hilderbrandt TB, Goeritz F, East ML. 2002. Movement of elephants in the Selous–Niassa wildlife corridor, southern Tanzania. Pachyderm 33:8–31. Naughton-Treves L. 1997. Farming the forest edge: vulnerable places and people around Kibale National Park, Uganda. Geographical Review 57(1):27–46. [NCRC] Nature Conservation Research Centre. 1999. The micro-best practices project: report of the ecological baseline survey in the Red Volta River Valley of the Upper East Region. Unpublished report. NCRC/ TechnoServe Inc., Accra. [NCRC] Nature Conservation Research Centre. 2000. The micro-best practices project: monitoring report on the ecological impact of the project for the second period. Unpublished project report. TechnoServe Inc., Accra. Okoumassou K, Barnes RFW, Sam M. 1998. The distribution of elephants in north-eastern Ghana and northern Togo. Pachyderm 26:52–60. Osborn FV. 2003. Seasonal influence of rainfall and crops on home-range expansion by bull elephants. Pachyderm 35:53–59. Osborn FV, Parker GE. 2002. Community-based methods to reduce crop loss to elephants: experiments in the communal lands of Zimbabwe. Pachyderm 33:32–38. Parker GE, Osborn FV. 2001. Dual-season crop damage by elephants in northern Zimbabwe. Pachyderm 30:49–56. Sam MK. 1994. A preliminary survey of elephants in northern Ghana. Unpublished report. Ghana Wildlife Depart- 48 ment, Accra. Sam MK, Ayensu S, Agbenu V, Kumordzi BB, Wilson S. 2003. Reconnaissance survey of human–elephant conflict in the Dadieso area, western Ghana. Pachyderm 35:132–135. Sam MK, RFW Barnes, Okoumassou K. 1998. Elephants, human ecology and environmental degradation in north eastern Ghana and northern Togo. Pachyderm 26:61– 68. Sam MK, Haizel C, Barnes RFW. 1997. Crop-raiding by elephants during the 1996 harvest season in the Red Volta Valley (Upper East Region Ghana). Unpublished report. Ghana Wildlife Department, Accra. Sam MK, Haizel CAK, Barnes RFW. 2002. Do cattle determine elephant distribution in the Red Volta Valley of northern Ghana? Pachyderm 33:39–42. Sawadogo B. 2003. Etude et cartographie des mouvements et des zones de conflicts des éléphants entre le Parc Nationale de Po dit Kabore Tambi et la frontière du Ghana le long de la Volta rouge. Unpublished report. IUCN/AfESG, Nairobi. Sukumar R. 1990. Ecology of the Asian elephant in southern India. 2. Feeding habits and raiding patterns. Journal of Tropical Ecology 6:33–53. Sukumar R. 1994. Wildlife–human conflict in India: an ecological and social perspective. In: Guha H, ed., Social ecology. Oxford University Press, New Delhi. p. 303–317. Sukumar R. 2003. Living elephants: evolutionary ecology, behavior and conservation. Oxford University Press, New York. Taylor CJ. 1952. The vegetation zone of Gold Coast. Bulletin of the Gold Cost Forestry Department 4:1– 2. Thouless CR, Sakwa J. 1995. Shocking elephants: fencing and crop raiders in Laikipia District, Kenya. Biological Conservation 72:99–107. Wheelan JH. 1950. Red Volta West Reserve working plan. Unpublished report. Forestry Services Division, Bolgatanga, Ghana. Pachyderm No. 38 January–June 2005 Nature and extent of human–elephant conflict in Bia Conservation Area, Ghana Nature and extent of human–elephant conflict in Bia Conservation Area, Ghana Moses Kofi Sam,1* Emmanuel Danquah,2 Samuel K. Oppong3 and Enoch A. Ashie4 1 Resource Management Support Centre, Forestry Commission, PO Box 1457, Kumasi, Ghana; email: [email protected], *corresponding author 2 A Rocha Ghana, PO Box KN, Kaneshie, Accra, Ghana; email: [email protected] 3 Institute of Renewable Natural Resources, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana; email: [email protected] 4 Bia National Park, Wildlife Division of Forestry Commission, Sefwi-Asempanaye, Sefwi-Wiawso, Ghana Abstract An investigation into the nature and extent of human–elephant conflicts in and around Bia Conservation Area was carried out during the 2004 rainy season. This exercise was done by administering questionnaires to wildlife staff and local communities as well as by actual field measurement of damaged farms. There were 49 elephant cropdamage incidents involving 44 farms belonging to 36 farmers. Elephant crop damage was a serious problem in the conservation area, with farmers around the southern portions being the most affected. The number of raids increased with the proximity of a cluster of farms to the park boundary, and the number of crop types. The area under cultivation could influence the number of raids. There was almost 50% probability that if one’s farm was raided, about half of the crop would be destroyed. This was exacerbated by the fact that raiding targeted mature and good-quality crops. The high damage levels have resulted in continuous friction between farmers and conservationists since most farmers do not see any advantage in conserving elephants. Résumé On a mené une enquête sur la nature et l’étendue des conflits hommes–éléphants dans et autour de la Zone de Conservation de Bia pendant la saison des pluies de 2004. Cet exercice a été réalisé au moyen de questionnaires remis au personnel de la faune et aux communautés locales ainsi que par des mesures directes, sur le terrain des dégâts causés aux fermes. Il y a eu 49 incidents où des éléphants ont dévasté des récoltes, impliquant 44 fermes appartenant à 36 fermiers. Les dégâts causés par les éléphants aux récoltes sont un problème sérieux dans l’aire de conservation, les fermiers se trouvant aux environs des parties sud étant les plus affectés. Le nombre de raids augmentait avec la proximité du groupe de fermes par rapport au parc et avec le nombre de types de cultures. La zone cultivée pouvait influencer le nombre de raids. Il y avait presque 50 % de risques que, si une ferme était attaquée, près de la moitié des récoltes soit détruite. Ceci était aggravé du fait que les attaques visaient surtout des cultures arrivées à maturité et de bonne qualité. Le taux élevé de destruction a entraîné des frictions continues entre les fermiers et les protecteurs de l’environnement étant donné que la plupart des fermiers ne voient aucun avantage à la conservation des éléphants. Introduction Human–elephant conflict (HEC) is a problem that many parks and reserves across Africa experience. This problem is especially severe in West Africa, where isolated populations of elephants often live Pachyderm No. 38 January–June 2005 adjacent to areas of dense agriculture (Sukumar 1990; AfESG 1999). As human populations increase and elephant populations become more concentrated in isolated protected areas and remnant forest habitats, these conflicts are almost certain to escalate (Barnes et al. 1995), making this problem one of Africa’s most 49 Sam et al. challenging conservation issues (Hoare and du Toit 1999). Crop damage by elephants around Bia Conservation Area (BCA) is a serious multifaceted management problem that authorities of the Wildlife Division of Ghana face (Sam 2000). While the problem has been investigated extensively (Barnes et al. 2003) to identify the underlying causes, there have not been many systematic data-gathering attempts on this issue in the area for park management to understand and appreciate habitat requirements and the crop-raiding behaviour of BCA elephants. To study the nature and extent of the HEC situation and to help park management tackle the problem effectively, we conducted a social survey in some of the affected communities and took measurements on affected farms. Study area The Bia Conservation Area comprises Bia National Park (NP) in the north and its adjoining Bia Resource Reserve (RR) in the south. The two forests form a block of 306 km2 located in the moist evergreen and moist semi-deciduous forest zones of western Ghana. BCA lies between latitude 6°20′ to 6º40′ N and longitude 3º00′ to 3º10′ W, sandwiched between the Bia River and the border with Cote d’Ivoire (fig. 1). The area has an annual precipitation of between 1500 and 1750 mm (Hall and Swaine 1976) with two peaks, in June and October. Average monthly temperature in the area falls between 28 and 24°C with extremes of 34 and 18°C. The farming system is rain fed, with farming activities being undertaken throughout the year, resulting in year-round crop raids. Methods To understand the human–elephant situation around BCA, both the historical and the current crop-raiding situations were determined. This was done through questionnaires and conducting interviews with 42 randomly selected members of 11 randomly picked N 6° 40´ Bia National Park N 6° 35´ N 6° 25´ Bia Rive r Bia Resource Reserve N 6° 30´ COTE D'IVOIRE Ghana N 6° 20´ W 3° 10´ W 3° 05´ W 3° 00´ 0 5 10 km Figure 1. Bia Conservation Area, Ghana. 50 Pachyderm No. 38 January–June 2005 Nature and extent of human–elephant conflict in Bia Conservation Area, Ghana fringe communities out of a total of 20 major com- of subjectivity in determining the quality scoring by munities within 7 km of BCA. We tried to determine defining the quality of various crops grown in the area the types of conflicts that occurred, how long each as poor, medium, or good and by ensuring that the had been going on, the frequency, spatial extent and same set of enumerators was used throughout the so on, through a questionnaire specifically developed study. for this study. With this approach, we gathered some We recorded the geographical coordinates of raided qualitative historical and current information on the farms with a GPS. By plotting relative positions on a distribution and frequency of crop raids around the map of the study area, we determined the distance of study area (Sam et al. 2003). raided farms from the nearest forest boundary. Information on current crop-damage incidents (usually gathered within 48 hours), crops raided, Results growth stage at which crops were raided, crops spared, and time of year raids occurred was gathered using Crop raiding in the study area is a serious problem, an elephant damage report form developed by the and it occurs throughout the year. Forty-two farmers, IUCN African Elephant Specialist Group (AfESG) aged 21 to 50 years, were arbitrarily drawn from the (Hoare 1999). The area in square metres of a raided 11 randomly selected communities (fig. 2) around farm was estimated by roughly subdividing the farm BCA and interviewed. Immigrants formed the mainto measurable shapes (squares, rectangles, triangles, jority (72%), while natives constituted the remaining etc.) and summing up the calculated areas. The total 28%. Forty-two farmers were interviewed, the majoraffected area of damaged crops was measured likeity of whom (57%) had not seen or had any physical wise. Extent of damage was scored for raided crops and farms based on an index of damage developed Kwakudua by Hoare (1999). Damage Asuopri N 6°40´ score was then scaled from 0 to >9 (integers only). Bia Kramokrom Manukrom Scores ≤5 were interpreted National Adjuofia Park as low and non-severe, N 6°35´ scores from 6–8 were ranked medium damage, Nkrumahkrom Bia and scores ≥9 were inter- N 6°30´ Resource Reserve preted as being high and severe. The damage score was N 6°25´ Sabin Akatiso the sum of the age score of crops (1 = seedling, 2 = Anwiafutu intermediate, 3 = mature), N 6°20´ Anwiafutu jct the quality score (1 = poor, Alhajikrom 2 = medium, 3 = good) and the damage category (1 = ≤5% of farm area damW 3°15´ W 3°10´ W 3°05´ W 3°00´ W 2°55´ W 2°50´ aged, 2 = 6–10 %, 3 = 11– 20%, 4 = 21–50%, 5 = Legend 0 5 10 km 51–80%, and 6 = > 80% = crop-raiding sites of farm damaged). With = communities interviewed our experience regarding the farming system in the Figure 2. Distribution of the study communities and crop-raiding incidents near area, we reduced the level Bia Conservation Area. Pachyderm No. 38 January–June 2005 51 Sam et al. encounter with elephants within the past six years. They also had no idea whether elephant numbers had increased or not. Most of the farmers (90%) employed different kinds of traditional deterrent methods for driving elephants away from their farms: noise making by beating on metal objects, and firing guns and carbide bombs were the most frequent. However, noise making alone was not very effective unless combined with other methods like burning car tyres or setting up fires during the night. It must be noted that in the wake of all these traditional ways of deterring elephants, most farmers would fear for their lives were they to come face to face with elephants on their farms, and hence they had always relied on the Wildlife Division guards to drive raiding elephants back to the reserves. Based on visual and track identification in eight different settings or occasions, it can be said that at least 24 males and 12 females were involved in the raids. In terms of age, some 43 adults and 33 subadults and infants had been seen on different occasions. Some of these could be the same elephants showing up in different places at different times. Severe crop damage starts in June and increases steadily before peaking in September and October. It declines in November and by December has become minimal. During the 2004 major farming season, 44 farms belonging to 36 farmers from 18 villages experienced 49 raids around the conservation area (table 1). Farmers whose farms border the south and south-eastern boundary line of Bia RR experienced the highest number of raids (fig. 2). There was no significant (NS) relationship between the number of raids and the size of individual farms raided (r2 = 0.057, NS) or the nearest distances of individual farms to the reserve boundary line (r2 = 0.102, NS). Hence the data were further analysed at two levels by regression. First, we examined cropraiding incidents for a particular area, that is, at the village level. For this level of analysis, the data for all raids within a common village were combined and related to the total cultivated (farmed) area of the village (table 1). The number of raids that a raided farm suffered was evaluated in relation to the area of land under cultivation in that area. Secondly, the data were analysed by relating the total number of raids in a particular village to the mean distance of raided farms in the village from the nearest reserve boundary. The number of raids registered in an area was inversely influenced (r2 = 0.857, p < 0.05) by the mean of their distances to the nearest reserve boundary line (fig. 3). For farms that were raided, the risk of a farm suffering damage increased with the total area cultivated Table 1. Crop-raiding incidences in relation to proportion of farm area destroyed around affected villages Villages Akosua Aden Krom Alhaji Nkwanta Anwiefutu Nkwanta Asiri Bio Krom Boampong Krom Camp 4 Camp 10 Eye Nyame Krom Iron Boy Kofiko Krom Kofi Kyere (Camp10) Kojo Donkor Camp Kwaku Boakye (Camp10) Kwasi Donkor Camp Kwasi Donkor Krom Nyamebekyere Yamediagoro Total 52 Mean Farmers distance to affected reserve (m) 2.5 3 1 3 2.5 4 3 3 2 2 3.5 4 1.5 3.5 2 1.5 2 0.8 Farms raided Total area Total area of farms in destroyed (m2) a village (m2) Raids registered 2 1 5 1 3 1 1 1 2 3 1 1 4 1 1 2 3 3 3 1 7 1 3 1 1 1 2 3 1 1 5 1 1 3 3 6 4,880 1,620 16,500 3,575 11,720 3,360 1,240 1,560 3,460 5,200 2,570 2,750 13,600 1,430 2,200 6,430 5,800 13,300 1,890 1,460 11,050 920 3,600 480 980 960 1,650 3,220 1,220 640 4,200 520 1,760 2,860 2,800 6,250 3 1 8 1 3 1 2 1 2 3 1 1 5 1 2 4 4 6 36 44 101,195 46,460 49 Pachyderm No. 38 January–June 2005 Nature and extent of human–elephant conflict in Bia Conservation Area, Ghana in a village farm enclave (r2 = 0.755, p < 0.05) (fig. 4). The risk of crop raiding also increased with the number of food crops on any individual raided farm (r2 = 0.756, p < 0.05) (fig. 5), that is, the more the In [number of raids +1] In [number of raids +1] different types of crops the higher the number of raiding incidents. Thus a farmer who monocropped was at low risk of elephant crop raiding. By planting two or more crops the farmer increased the risk. There was no significant relationship between raids and acreage 2.4 of individual crops grown: cas2.2 sava, r2 = 0.160, NS; plantain, r2 = 0.604, NS; cocoa, r2 = 0.011, NS; 2 maize, r2 = 0.507, NS; yam, r2 = 1.8 0.044, NS; cocoyam, r2 = 0.541, NS; banana, r2 = 0.063, NS and 1.6 vegetables, r2 = 0.063, NS. Never1.4 theless, for five of the eight crops cultivated in the area, there seems 1.2 to be an unusual phenomenon (ta1 ble 2)—a U-shaped relationship between number of crop-raiding 0.8 incidents and sizes of crops cultivated, suggesting that when crops 0.6 6.6 6.8 7 7.2 7.4 7.6 7.8 8 8.2 8.4 are grown in modest amounts they are raided least or not at all. In [distance to reserve (m) +1] Farming in the study area Y = 1.609 + .917 * X – .126 * X^2; R^2 = .857 involved seasonal, rain-fed subsistence agriculture. Cassava was Figure 3. Relationship between number of raids registered in an area the most raided crop (30% of crops and the mean distance of farms in a village from the nearest reserve raided), followed by plantain boundary line. (26%) (table 3). Raiding was largely targeted at crops that were 2.4 mostly mature (71% of total fre2.2 quency of crops raided). Crop raiding was largely targeted at crops 2 that were of good quality (69% of total frequency of crops raided) (ta1.8 ble 3). Table 4 also indicates that 1.6 most crops other than cocoyam suffered an appreciable level of dam1.4 age. 1.2 Damage on half the raided farms amounted to about 46% with 1 about 5% suffering more than 80% 0.8 damage (fig. 6). 0.6 7 7.5 8 8.5 9 9.5 10 In [area cultivated (sq m) +1] Y = 10.651 – 2.758 * X + .193 * X^2; R^2 = .755 Figure 4. Relationship between number of raids registered in an area and the area of land under cultivation (farm). Pachyderm No. 38 January–June 2005 Discussion Elephant crop raiding in BCA is a serious problem that occurs throughout the year. It dates back into the 1970s when immigrant farmers 53 Sam et al. 2.4 In [raids registered +1] 2.2 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 In [number of crop species +1] Y = 1.114 * X + .957 * X^2; R^2 = .756 Figure 5. Relationship between number of crop species grown and number of raids registered in that area. started cultivating between reserves (pers. comm. Phillip Mensah, Camp 9 leader, February 2004). Although there are no data to show the trend in crop-raiding frequency over the last two or three decades, there is much anecdotal evidence supporting an increasing trend with a growing influx of migrant farmers. Consequently, the problem has developed into a big issue, and the Wildlife Division is constantly under pressure from local communities to curb it. Wildlife guards are blamed by irate farmers for their inability to control the elephants. Based on data gathered through the questionnaire, it appears that elephant behaviour has changed; according to farmers, in the past it was primarily males that raided and then Table 2. Frequency (f) and percentage of crop raiding in relation to the area of each species grown on each farm Cultivated area 2 2 0m Cassava Plantain Cocoa Maize Yam Cocoyam Banana Vegetables 1000–1999 m2 1–999 m 2000–2999 m2 > 3000 m2 f % f % f % f % f % 10 26 45 11 45 30 48 48 20 53 92 22 92 61 98 98 0 0 0 0 4 0 0 1 0 0 0 0 8 0 0 2 6 4 0 3 0 5 1 0 12 8 0 6 0 10 2 0 9 0 1 4 0 14 0 0 18 0 2 8 0 29 0 0 24 19 3 28 0 0 0 0 49 39 6 57 0 0 0 0 Table 3. Stage of growth of crops damaged on 44 farms raided by elephants, 2004 Crop How many times crops were raided Percentage of raided crop damaged Crops in mature stage (no.) Crops in intermediate stage (no.) Crops of good quality (no.) Crops of medium quality (no.) Crops of poor quality (no.) Cassava Plantain Cocoa Maize Yam Cocoyam Banana Vegetables 25 21 15 13 6 1 1 1 30 26 18 16 7 1 1 1 15 18 11 12 2 0 0 1 10 3 4 1 4 1 1 0 11 16 11 12 5 0 1 1 7 2 2 1 1 1 0 0 7 3 2 0 0 0 0 0 Total 83 59 24 57 14 12 No crops in the seedling stage were reported raided 54 Pachyderm No. 38 January–June 2005 Nature and extent of human–elephant conflict in Bia Conservation Area, Ghana Table 4. Damage score for crops raided by elephants Crop Cassava Plantain Cocoa Maize Yam Cocoyam Banana Vegetables Age score Quality score Damage category Damage score 3 3 3 3 2 2 2 3 2 3 3 3 3 2 3 3 5 4 2 5 4 1 4 1 10 10 8 11 9 5 9 7 Interpretation high and severe high and severe medium high and severe high and severe low and not severe high and severe medium 14 Number of farms 12 10 8 6 4 2 0 0–10 11–20 21–39 31–40 41–50 51–60 61–70 71–80 81–90 91–100 Percentage of damage Figure 6. Frequency distribution by percentage of area damaged on raided farms. only at night. In recent times, however, family groups consisting of adult females, males and especially subadults are often the culprits. Elephants have also been seen in the fields in broad daylight. This is also true at the Kakum CA, and to help the Wildlife Division plan a more effective crop-raiding deterrent method, a concerted investigation should be made into this change in elephant behaviour at the two sites. While damage may be restricted to the wettest part of the year, this study recorded most damage in September and October coinciding with the minor rainy season. In the Kakum CA, severe damage occurs in June, coinciding with the major rainy season (Dudley et al. 1992). Both Barnes et al. (2003) and Danquah (2003) discuss crop raiding in relation to rainfall at Kakum CA . Damage in the Red Volta area is severest in October, when the single rainy season would be ending and most crops would be being harvested (Sam 2000). Pachyderm No. 38 January–June 2005 Elephants were originally found in both Bia NP and Bia RR (Martin 1982; Short 1982). Since timber companies started logging in Bia RR in the early 80s (Parren et al. 2002) elephants have left Bia NP and moved downwards into the south-eastern portions of Bia RR. Thus it has been suggested that the absence of elephants in Bia NP was temporarily a reaction to the different and more palatable secondary vegetation conditions created by logging within Bia RR (Short 1981; Martin 1982). Both Barnes (1996) and de Leede (1994) observed a similar pattern where more elephants were observed in the south-west of BCA. Occurrences of crop raids have been reported mainly in the wet season, along the eastern borderline of Bia RR where elephants concentrated (Opoku 1988; de Leede 1994). Moreover, Martin (1982) has indicated that the Bia population usually confined movements to the same, often traditional, routes. 55 Sam et al. However, present distribution is gradually changing; records of elephant activities probably dating back to the previous rainy season around the northern boundary lines of Bia RR indicate that elephant movements and crop raiding, in addition to concentrating in the south and south-eastern Bia RR boundary line, occasionally spread northwards. Similarly, reports from wildlife staff and local communities indicate that elephant crop raiding actually spreads periodically to the northern sections including areas adjacent to Bia NP. It is believed that this pattern occurs during the late rainy crop-growing season when water sources increase throughout both reserves, and food crops like maize mature around the park boundaries. These two factors may be the most important determinants of elephant distribution in the wet season. However, a section of the park staff still contends that the relative increase and spread in raiding activities is a result of elephants crossing over from neighbouring Cote d’Ivoire during the wet season. The strange U-shape relationship between raids and abundance of certain raided crops is difficult to explain—the incidence of raids ranged from low to high when a farm had little or none of specific crops in an area. In moderate amounts (1–2999 m2), crop raiding fell to almost zero. Then the frequency of raids rose again with larger amounts (> 3000 m2) of these crops. A similar relationship was reported at Kakum CA for cocoyam (Barnes et al. 2003). However, within the limits of data gathered for raided areas, the most consistent lesson here and advice to farmers is that elephants may avoid a modest-size farm. At this stage we cannot explain why this should be so. The pattern of raiding suggests that elephants usually raid farms clustered close to the park, and for those farms raided, the area under cultivation and the number of different crop types were major predictors of raiding. The mean distance from the boundary line was the strongest predictor of risk, and the same was true for Kakum CA (Barnes et al. 2003) in Ghana and of Kibale NP in Uganda (Naughton-Treves 1998). Given that people must eat, and that the current policy of the government of Ghana is to conserve the country’s last remaining elephants, we need to search for a form of agricultural practice that reduces the risks of attracting elephants. Cultivation of food crops should be discouraged within the immediate environs of the reserves. Hence, the most effective action a farmer can take is to move away from the park boundary. If a farmer is incapable of resettling and farming else- 56 where, then that farmer has to reduce the types of crops grown or cultivate crops in modest amounts. Barnes et al. (2003) made this recommendation for Kakum CA as well. Sam et al. (1997) recorded two types of food crop damage. First is damage elephants cause by walking across farmland without feeding extensively, referred to as ‘collateral damage’, as in military parlance. The second type is when elephants intentionally stop to feed on crops. In this case, the percentage of crop damage is high and can be of real economic consequence to the farmer. While Sam et al. (1997) recorded damage of less than 10% in most fields in the Red Volta area in northern Ghana and only about 20% of the second type, the situation at Bia CA was the far different; only 2% of farms suffered collateral damage, and over 85% suffered between 21 and 70% of damage. Although Martin in 1982 said that damage caused by elephants to farms might be completely negligible for communities around Bia CA, Sam (2000) recorded the severest farm damage around BCA as being 40%. The current range of 14 to 93% suggests that crop raiding is becoming more severe. Single bulls and bull groups as well as family groups are involved. Farmers should be encouraged to protect their crops. Protection comes in two stages: detection and repulsion (Osborn and Parker 2002; Barnes et al. 2003). Improving methods of detecting the approach of elephants can considerably reduce the chance of damage (Osborn and Parker 2002). But at present, most farmers detect elephants only when they are already on the farm. For this reason farmers mostly concentrate their ability on repelling elephants, trying to send them back into the park. Unfortunately, elephants quickly habituate to any one method of repulsion; hence to successfully drive elephants away a number of methods in combination are necessary. Farmers interviewed confirmed that they must always combine noise making with other scaring tactics to be successful in driving elephants away from their fields. Consequently, farmers spend much time, resources and money to mitigate these conflicts. Under such conditions, villagers often resort to many forms of violence. The fact that farmers take such risks, in addition to the other problems of farming near the park (limited road access, distance from the village), demonstrates the intense demand for land in the area. It also means that subsistence farming is not a suitable form of land use around BCA, a situation also observed around Kakum CA (Barnes et al. 2003). Pachyderm No. 38 January–June 2005 Nature and extent of human–elephant conflict in Bia Conservation Area, Ghana This paper’s weakness is that we looked only at farms that were raided. As we did not look at those not raided we cannot calculate the percentage of farms that were affected by this problem to show its gravity. Also we cannot calculate what proportion of the farms growing cassava, plantain, and so on were raided. Knowledge of the features of the farms that were not raided, that is the successful farms, and why they were successful, would have helped us realize what raided farm owners need to do to move their farms into the ‘undamaged’ category. Acknowledgements We wish to thank the African Elephant Specialist Group through its EU-funded Small Grants Programme and A. Rocha Ghana for financing this study. We also acknowledge the contributions of Eben Daryl Bosu, Bright Kumordzi, Frank Tetteh Kumah, Aba Odoi-Agyarko and Sakyibea Biney (the BP 2002 Award Winning Team from Kwame Nkrumah University of Science and Technology) for their untiring support of the project. Our special thanks go to Dr Richard Barnes for reading through the script. Lastly, we are thankful to the entire staff of the Bia Conservation Area, especially the senior wildlife officer in charge, his deputy, Senior Wildlife Protection Officer Boamab, and Ranger Prince Charles Asante for providing staff and other facilities. References [AfESG] African Elephant Specialist Group. 1999. Strategy for the conservation of West African elephants. Proceedings of a workshop held in Abidjan, 22–26 February 1999. AfESG, Ouagadougu. Unpublished. Barnes RFW. 1996. Training course in elephant biology for Ghanaian wildlife officers. Report to US Fish and Wildlife Service. Unpublished. Barnes RFW, Azika S, Asamoah-Boateng B. 1995. Timber, cocoa and crop-raiding elephants: a preliminary study from southern Ghana. Pachyderm 19:33–38. Barnes RFW, Barnes KL, Alers MPT, Bloom A. 1991. Man determines the distribution of elephants in the rainforests of north-eastern Gabon. African Journal of Ecology 29:54–63. Barnes RFW, Boafo Y, Nandjui A, Dubiure UF, Hema EM, Danquah E, Manford M. 2003. An overview of crop raiding by elephants around the Kakum Conservation Area. Parts 1 and 2. Elephant Biology and Manage- Pachyderm No. 38 January–June 2005 ment Project, Africa Program, Conservation International. Unpublished. Danquah E. 2003. Feeding behaviour of the forest elephant and logging impact on fruit production in the Kakum Conservation Area. M. Phil. dissertation, Kwame Nkrumah University of Science and Technology, Kumasi. Unpublished. de Leede BM. 1994. Feasibility study on the establishment of corridors for forest elephants (Loxodonta africana cyclotis, Matschie) between forest reserves in western Ghana and eastern Cote d’Ivoire. Ghana Wildlife Department, Accra, Ghana. Unpublished. Dudley JP, Mensah-Ntiamoah AY, Kpelle DG. 1992. Forest elephants in a rainforest fragment: preliminary findings from a wildlife conservation project in southern Ghana. African Journal of Ecology 30:116–126. Hall JB, Swaine MD. 1976. Classification and ecology of closed canopy forest in Ghana. Journal of Ecology 64:913–951. Hoare RE. 1999. Data collection and analysis protocol for human–elephant conflict situations in Africa. Document prepared for the IUCN African Elephant Specialist Group’s Human–Elephant Conflict Working Group, Nairobi. Hoare RE, du Toit JT. 1999. Coexistence between people and elephants in African savannas. Conservation Biology 13(3):633–639. Martin C. 1982. Management plan for the Bia Wildlife Conservation Areas. General part (1) and final report. IUCN/WWF project 1251. Unpublished. Naughton-Treves l. 1998. Predicting patterns of crop damage by wildlife around Kibale National Park, Uganda. Conservation Biology 12:156–168. Opoku GK. 1988. The elephant (Loxodonta africana cyclotis) farm raiding around the Bia National Park: incidents, causes and solutions. BSc thesis. Institute of Renewable Natural Resources, University of Science and Technology, Kumasi. Unpublished. Osborn FV, Parker GE. 2002. Community-based methods to reduce crop loss to elephants: experiments in the communal lands of Zimbabwe. Pachyderm 33:32–38. Parren MPE, de Leede BM, Bongers F. 2002. A proposal for a translational forest network area for elephants in Cote d’Ivoire and Ghana. Oryx 36(3):249–256. Sam MK. 2000. The distribution of elephants in relation to crop damage around Bia Conservation Area during the 1999 rainy season. Report for IUCN (World Conservation Union), Gland, Switzerland. Unpublished. Sam MK, Haizel C, Barnes RFW. 1997. Crop raiding by elephants during the 1996 harvest season in the Red 57 Sam et al. Volta Valley (Upper East Region, Ghana). WWF Project 9F0062, Wildlife Department, Accra, and University of California at San Diego. Unpublished. Sam MK, Ayesu S, Agbenu V, Kumordzi BB, Wilson S. 2003. Reconnaissance survey of human–elephant conflict in the Dadieso area, western Ghana. Pachyderm 35:132–136. Short JC. 1981. Diet and feeding behaviour of the forest elephant. Mammalia 45:178–185. 58 Short JC. 1983. Density and seasonal movements of the forest elephant (Loxodonta africana cyclotis, Matschie) in the Bia National Park, Ghana. African Journal of Ecology 21:175–184. Sukumar R. 1990. Ecology of the Asian elephant in southern India. 2: Feeding habits and crop-raiding patterns. Journal of Tropical Ecology 6:33–53. Pachyderm No. 38 January–June 2005 Population survey of elephants in Cross River National Park, Nigeria Population survey of elephants in Okwangwo Division, Cross River National Park, Nigeria Emmanuel Obot,1* Clement Edet,2 Gabriel Ogar3 and Joy Ayuk2 1 Nigerian Conservation Foundation, Lekki Conservation Centre, PO Box 74638, Victoria lsland, Lagos, Nigeria; email: [email protected]; *corresponding author 2 Cross River National Park, Okwangwo Division, PO Box 1028, Calabar, Nigeria 3 Living Earth Nigeria Foundation, PO Box 8060, Port Harcourt, Nigeria Abstract Between March 1997 and May 1998, a census of elephant populations using dung counts was carried out along five different trails constructed in Cross River National Park, Okwangwo Division, Nigeria. The survey covered 139 km of trails. Estimated elephant density was 0.3 elephants per km2 with a variation of 0.2– 0.5 elephants per km2. This study is the first record of elephant population densities since the area became a national park in 1991. There appears to be a seasonal local migration of this population between Cross River National Park and the neighbouring Takamanda Forest Reserve in Cameroon; the elephants spend most of the rainy season in Cameroon and the dry season in Cross River National Park. Reasons to explain this migration were not apparent during this survey. Protection of this population will depend on closer cooperation with authorities in the Takamanda Forest Reserve. Résumé Entre mars 1997 et mai 1998, on a réalisé un recensement des populations d’éléphants en se servant du comptage des crottes le long de cinq pistes différentes construites dans le Parc National de Cross River, Division d’Okwangwo en Nigeria. L’étude se faisait sur 139 km de piste. La densité d’éléphants fut estimée à 0,3 / km2, avec une variation comprise entre 0,2 et 0,5 éléphants au km2. Cette étude est le premier rapport sur la densité des éléphants depuis que la région est devenue un parc national, en 1991. Il semble qu’il y ait une migration saisonnière de la population entre le Parc National et la Réserve Forestière voisine de Takamanda, au Cameroun. Les éléphants passent la plus grande partie de la saison des pluies au Cameroun et la saison sèche dans le Parc National de Cross River. Les raisons de cette migration n’ont pas été perçues lors de cette étude. La protection de cette population dépendra de la coopération plus étroite avec les autorités dans la Réserve forestière de Takamanda. Introduction Nigeria has lost 96% of its original lowland rainforest and consequently a considerable percentage of forestdwelling and -dependent fauna. In 1991 Cross River National Park was created to protect the remaining 4% of relatively undisturbed rainforest. This report presents the results of the first attempt to estimate elephant (Loxodonta africana cyclotis) populations in the park. Study area Cross River National Park is made up of two discontinuous sectors—the southern Oban Hills Division Pachyderm No. 38 January–June 2005 and the northern Okwangwo Division. This study was carried out in Okwangwo Division, which lies between 6°4' and 6°29' N and 9° and 9°27' E. Together with the Mbe Mountains it covers 920 km2, lying south-east of Obudu with the eastern boundary extending along the Cameroon border. It is part of the Guinea–Congolia–Sudania Regional Transition Zone with an afromontane archipelago-like centre of endemism in Obudu Plateau and the Sankwala Mountains (figs. 1, 2). With the contiguous Afi River Forest Reserve to the west and Takamanda Forest Reserve (Cameroon) to the east, the Okwangwo Division of Cross River National Park lies within the core of a continuous 59 Obot et al. 8º00' E 8º30' E 8º30' E 9º00' E N I G E R I 6º30' N A OKWANGWO DIVISION Cross River National Park AFI RIVER Forest Reserve TAKAMANGA Forest Reserve Cros s Riv er 6º00' N Ikom N I G E R I A Mamfe OBAN HILLS DIVISION Cross River National Park 5º30' N Cr os sR ive r OBAN HILLS DIVISION Cross River National Park Nguti KORUP National Park C A M E R O O N 5º00' N Calabar r ive sR os Cr 0 10 20 Mundemba Kumba 6º30' N 30 km Atlantic Ocean international boundary roads Figure 1. Location of Cross River National Park. 60 Pachyderm No. 38 January–June 2005 Population survey of elephants in Cross River National Park, Nigeria MB EM OU NT AIN S moist tropical rainforest system of more than 1800 the dense vegetation, rugged terrain, low density, and km2 that is completely or partly protected. rather patchy distribution of elephants. Okwangwo Topography is generally rugged, with many disjunct Division is within the same lowland rainforest comand connected ridge systems, isolated peaks and rock plex with a similar vegetation structure. Moreover, outcrops. Land, generally at elevations of 150 m, rises preliminary recognizance studies suggested a lowto around 1500 m in the Sankwala Mountains and to density, highly mobile population. In this study, thereodd peaks of up to 1700 m on the Obudu Plateau. Three fore, human trails were regarded as adequate transects main rivers, Oyi, Bemi and Okon—all tributaries of because elephants in the study area seem to search the Cross River—drain the park area (fig. 2). out hunter and gatherer camps, where they eat fruit The climate is tropical with a distinct rainy season of the bush mango (Irvingia sp.), gathered and procbetween March and November, and a dry season be- essed in such camps. tween December and February. Rainfall is heavy: up to All records of sightings of elephants and their ac4280 mm distributed unevenly within the nine-month tivities in the park area had occurred in the central rainy season. Ambient temperatures are high but lower lowland area in Okwangwo in the north, Bamba in temperatures (14–16°C daily minimums and 18–25°C the east and Bashu on the southern border of the park daily maximums) are recorded on the highland areas (figs. 2, 3). This area, approximately 239 km2, was of Obudu Plateau and the Sankwala Mountains. A detailed account of the flora of SANKWALA MOUNTAINS Okwangwo Division is given in Bakle Obot (1996), who segregated the vegetation into four major types: lowland rainforest in Ongwelua Okwabang low-lying areas, ridge and hill Yangwape forest on the slopes of the Buobre Mbe Mountains, submontane Ochakwe1 Bulalang forest on the Obudu Plateau, Wula 1 and savanna woodland in the Wula 2 Ikwete Hills. Soils in the lowOBUDU land area are heavily leached, Balegete CATTLE RANCH extremely infertile and ferrallitic while in the highlands Okwa 1 they are generally ferruginous Okwangwo and susceptible to erosion. For Okwa 2 Bamba specific accounts of the Kanyang 2 Kanyang 1 Obudu Plateau and the Sankwala Mountains see Hall and Medler (1975), Medler and Hall (1975) and Keay 0 3 15 km (1979). Scale 1 : 25,000 Method of study Abo Mapang Obanyi Census In an earlier survey of elephant populations in the Oban Hills sector of the park, Dickinson (1995) found the line-transect method impractical in view of Bashu Okpambe international boundary park boundary main road minor road track river human settlement Figure 2. Cross River National Park, Okwangwo Division, with the proposed Mbe Mountains extension and the Obudu Cattle Ranch. Pachyderm No. 38 January–June 2005 61 Obot et al. determined by counting 1-km2 grids within which elephants had been sighted or their activities recorded since November 1994. Randomly placed transects were started from known landmarks—such as village, river, rock formation—and terminated at predetermined landmarks such as known camps frequently used by gatherers. These landmarks were georeferenced using a portable Garmin 40 Global Positioning System (GPS) where possible. (The Garmin GPS 40 does not operate properly under heavy tree canopy.) Several GPS points for mapping were also taken along the transect. Elephant dung piles of all ages were counted within a fixed width of 5 m on both sides along each transect. where D is dung density with a dung decay rate of 0.0462 (Dickinson 1995) and a defecation rate per elephant of 20 dung piles per day (Tchamba 1992). Results and discussion Table 1 summarizes data collected during this survey. Within the park elephants occupied a home range of approximately 239 km2 from Okwangwo in the centre of the park through Bamba on the eastern park boundary to Bashu in the south (figs. 2 and 3). The distance covered in the five transects was 139 km. The mean length of transects was 27.8 km. Mean dung density was 134.75 piles per km2. The extrapolated elephant density is therefore 0.3 elephants per km2. This adds to 74 elephants for the home range of approximately 239 km 2. The lowest individual transect estimate was 47 animals in transect 4, the highest 130 animals in transect 3. There are perhaps anywhere between 50 and 130 animals. In Okwangwo Division interviews with hunters and farmer-gatherers suggest that there are 10 groups of elephants with 15 individuals per group. These add up to 150 animals within the home range. This number compares favourably with our estimate of a maximum of 130 elephants. Albeit so, this is a very small population. Elephant hunting for tusks and for meat was extremely intense in the Okwangwo area before the national park was created. This slaughter has been controlled by ongoing anti-poaching patrols, and these anti-poaching activities must be sustained if this small population is to survive. Density One method of estimating dung densities is by extrapolating from linear correlations between actual dung density and the number of 0.5-km segments along a transect in which dung was recorded (Fay and Agnagna 1991). The relationship was: D = 9.688 + 25.016p (r = 0.804) where D is dung density and p is the number of 0.5km segments containing dung. To extrapolate elephant density from dung density, a conversion factor of 0.00231 was used based on a dung decay rate of 0.0462 estimated for the Oban Hills sector of the park (Dickinson 1995) and a defecation rate per elephant of 20 dung piles per day, estimated for Cameroon (Tchamba 1992). Thus, elephant density E is given by E = D(0.0462/20) elephants per km2 Table 1. Summary of dung count data along transects Transect no. 1 2 2 4 5 Totals Mean Significance 62 Length of transect (km) 58 15 43 14 9 139 27.8 21.51 Segments, Segments Dung 0.5-km with dung density (no.) 0.5-km (no.) (km2) 116 30 86 28 18 278 5 4 9 3 4 25 Elephant density (km2) 134.8 109.7 234.8 84.7 109.7 0.31 0.25 0.54 0.20 0.25 134.75 58.67 0.311 0.136 Estimated Sampling period no. of elephants 74 61 130 47 61 rainy season rainy season dry season dry season dry season 74.39 32.39 Pachyderm No. 38 January–June 2005 Population survey of elephants in Cross River National Park, Nigeria Elephant distribution Elephant distribution within the park, indicated with 1km2 blocks around GPS points where elephant activities were recorded, is shown in figure 3. The elephants occupy the central lowland area of the park between Okwangwo in the northern, Bamba in the eastern, and Bashu on the southern borders of the park (figs. 2, 3). There is a strong seasonal migration of the population between the park and Takamanda Forest Reserve, Cameroon (fig. 1) with elephants spending most of the rainy season (March–August) in Cameroon and the dry season in Okwangwo Division. The long-term survival of this population depends on protection efforts complementary with Takamanda. At present, however, Takamanda is managed as a forest reserve where limited hunting is permitted. While anti-poaching activities are being intensified in Cross River National Park, evidence exists showing that poachers wait in Takamanda for the elephants during the seasonal migration. This emphasizes the great need for cross-border cooperation between authorities in Takamanda and conservation agencies in Nigeria. Such cooperation will protect not only the elephant populations but the entire fauna and flora of the Okwangwo–Takamanda block of rainforest. References Dickinson B. 1995. A reconnaissance survey of elephant population in Oban Division of Cross River National Park, Nigeria. Cross River National Park, Akamkpa, Nigeria. Fay M, Agnagna M. 1991. A population survey of forest elephants (Loxodonta africana cyclotis) in northern Congo. African Journal of Ecology 29:177–187. Hall JB, Medler JA. 1975. Highland vegetation in southeastern Nigeria and its affinities. Vegetatio 29:191–198. Keay RWJ. 1979. A botanical study of the Obudu Plateau and Sankwala Mountains. Nigerian Field 44(314):106– 119. Medler JA, Hall JB. 1975. The flora of the Obudu Plateau and associated highlands: an annotated species’ list. Herbarium Bulletin No. 9. University of Ife, Ife, Nigeria. Obot EA. 1996. Flora and vegetation of Okwangwo Division, Cross River National Park. In: Essential partnership: the forest and the people, Proceedings of a workshop on the rainforest of south-eastern Nigeria and south-western Cameroon, Obudu Cattle Ranch and Resort, 20–24 October 1996. Tchamba MN. 1992. Defecation by the African forest elephant (Loxodonta africana cyclotis) in the Sanchou Reserve, Cameroon. Nature et Faune 7:27–31. Acknowledgements Figure 3. Elephant distribution, shown in 1-km2 blocks around GPS points where elephant activities were recorded, in Cross River National Park, Okwangwo Division. Pachyderm No. 38 January–June 2005 Funds for this survey were provided by the Cross River National Park Project supported by the European Union, the World Wide Fund for Nature-UK, the Department for International Development of the UK, the Federal Government of Nigeria, and the Nigerian Conservation Foundation. 63 Fischer Elephants in Cote d’Ivoire—a warning for West African conservation Frauke Fischer Zoology III, Dept. of Tropical Biology and Animal Ecology, Theodor Boveri Institute Biozentrum, Am Hubland 97074, Würzburg, Germany; email: [email protected] Abstract Cote d’Ivoire once hosted probably one of the largest elephant populations in West Africa. Despite early warnings, numbers have decreased rapidly within the last century, resulting in small, isolated patches of populations. While savanna elephants suffered more during peak times of the ivory trade, forest elephant numbers declined rapidly with the destruction of their habitat. It is most likely that elephant numbers in all habitats and sites dropped significantly during the last decade, making the long-term survival of elephants in Cote d’Ivoire questionable at the very least. Major elephant habitats in forest (Taï National Park) and savanna (Comoé National Park) remain intact, but potential immigration from other sites or even re-introduction of elephants would require a significant improvement in park management and nature conservation in Cote d’Ivoire. Additional key words: savanna elephants, forest elephants, population trends, elephant conservation Résumé La Cote d’Ivoire a probablement hébergé dans le passé une des plus grandes populations d’éléphants de l’Afrique de l’Ouest. Malgré le fait que de nombreux signaux d’alarme aient été tirés, le nombre des éléphants a rapidement diminué pendant le siècle dernier, ayant pour résultat actuel des populations isolées et réduites. Les éléphants de savane ont beaucoup souffert lors des pics atteints par la commercialisation de l’ivoire tandis que les éléphants de forêt ont été affectés par la destruction de leur habitat. Il est fort probable que le nombre d’éléphants dans tous les habitats a diminué de manière significative durant les 10 dernières années ce qui remet très sérieusement en question la survie de ces animaux en Cote d’Ivoire au long terme. De grandes superficies de forêts (Parc National de Taï) et de grandes superficies de savanes (Parc National de Comoé) qui pourraient servir d’habitat pour les éléphants sont encore intactes, mais de grandes migrations dans ces régionslà ou des réintroductions ne seront pas possible sans la mise en œuvre d’efforts importants concernant la gestion des parcs nationaux et la protection de la nature en Cote d’Ivoire. Mots clé supplémentaires : éléphants de savane, éléphants de forêt, tendances de population, la protection des éléphants Introduction In pre-colonial times Cote d’Ivoire once hosted probably one of the largest elephant populations in West Africa. Even though no estimates of the pre-colonial elephant population exist, we can assume that due to the area’s vegetation structure, rainfall pattern and low human population density elephant numbers must have been tremendous. Applying Parker’s model (1989) and 64 with an assumed forest cover of roughly half of the country’s 322,460 km2, Cote d’Ivoire’s savanna elephant population alone might once have been 165,000 to well over 300,000 animals, depending on the human population density. However, pre-colonial empires existed, and wars, ivory trade and habitat destruction due to cultivation might have had a greater impact on West Afri- Pachyderm No. 38 January–June 2005 Elephants in Cote d’Ivoire can elephant populations than on those in eastern and southern Africa, even well before the arrival of the Europeans (Barnes 1999). Despite these early negative effects, elephant numbers remained high until colonial structures were established in West Africa, resulting in the name Ivory Coast—Cote d’Ivoire—given by the colonial powers to the area in West Africa that was richest in elephants (or at least elephant tusks). Although the name may reflect to a certain extent the importance of Cote d’Ivoire’s harbours for the international ivory trade, with part of the ivory coming from other areas, it hints of large elephant populations prevalent at that time. 19th century Between 1850 and 1875 ivory exports from Africa increased fourfold (Barnes 1999); 1836 and 1908 were peak years of ivory exports from French West Africa (Roth and Douglas-Hamilton 1991) with most ivory probably coming from the larger savanna elephants north of the forest belt. This led to a rapid decrease in savanna elephant populations with less effect on the forest elephants at that time (Roth and Douglas-Hamilton 1991). The decline of forest elephants accelerated with the conversion of forest to agricultural land and an ever-increasing destruction of forest habitats. 20th century Ivory exports from Africa peaked in the early 20th century, leading to a crash of the elephant population before World War I. Despite rising prices, ivory exports remained low, clearly indicating that elephants were being over-harvested as early as then (Roth and Douglas-Hamilton 1991). Roth and Douglas Hamilton (1991) estimated that during their study period between 1976 and 1984 elephants roamed over only 6–7% of their former range in West Africa, with savanna elephants suffering from greater range reduction than forest populations. Roth et al. (1984) estimated the total elephant population in Cote d’Ivoire to comprise not more than 1790 savanna and 3050 forest elephants scattered throughout the country within 46 more or less isolated populations. The annual loss at that time was approximately 300 poached and 90 legally killed forest elephants, resulting in an estimated annual decline of 10%. Numbers declined further by the late 1980s when Merz and Hoppe-Dominik (1989) presented data on the remaining 20 isolated populations of for- Pachyderm No. 38 January–June 2005 est elephants with a further 50% decline in numbers, and even further by the early 90s when the total number of both savanna and forest elephants had reached an all-time low of 63 to 360 elephants (speculative 666) in the entire country, existing in 24 populations (Blanc et al. 2003). Present situation Today the situation is probably even worse, with only seven elephant populations confirmed, comprising approximately 270 elephants (table 1). The outbreak of civil war in Cote d’Ivoire in September 2002 negatively affected the management of certain protected areas and conservation in general (Fischer 2004), and although not quantified it can be assumed that the effect on the small population of elephants was negative if not catastrophic. Cote d’Ivoire has eight national parks and 298 classified forests with the latter being sustainably managed (fig. 1). Within the last 50 years, however, large portions of classified forests have been clear-cut and converted to agricultural land. Very little intact forest habitat remains; it is largely in the east of the country in the Bossématié forest system and in Taï National Park (NP) in the south-west, which is treated as a protected area in this article. The forests are threatened by logging and agricultural activities, which will probably increase due to economic problems in Cote d’Ivoire, and the recent cessation of a long-term management and conservation programme funded by German development agencies. Six of the eight national parks, Azagny, Comoé, Marahoué, Mont Sangbé, Mont Péko and Taï, are known to host elephants, some in extremely low numbers (see below). Elephants occurred both within and outside protected areas until the 1980s (Roth et al. 1984) but are now more or less restricted to the larger national parks and some forest reserves (see ‘Elephants within protected areas’ under Results). As earlier reports on elephants, shot or alive, gave imprecise point locations it is not always clear whether the animals occurred inside or outside protected areas. For example, elephants shot near Abidjan (Roth et al. 1984) might have lived either in Banco NP or outside this protected area. The aim of this article is to contribute to the knowledge of the status of elephants in Cote d’Ivoire, giving the most updated information about this endangered, charismatic species in one of its former strongholds. I also raise additional awareness on the 65 66 1994 1989 1994 1991 2004 1997 2003 150 20–30 150 very few 30 30 1988/99 1994 200 55 60–80 1979/80 250 1978 1987 2003 1979/80 1990 1996 1977 1979 1979/80 1995 2002 1979/80 1983 1991 2002 1979 1979/80 1982 1989 2000 Year ground census (transects) informed guess informed guess informed guess informed guess informed guess informed guess ground census (transects) informed guess no details given aerial survey aerial survey line transects, dung counts no details given no details given other guess aerial survey aerial survey aerial survey informed guess informed guess no details given line transects from vehicle informed guess mark-recapture/faecal DNA no details given no details given ground census (transects) ground census (transects) Method 20 in Sogan-Tamin-Mabi-Yaja, 5 in Béki, 30 Bossématié Bossématié and Béki (5–6 elephants) CF Songan only (?) CF Songan only No elephants seen during reconnaissance flight Bossématié and Béki forests MIKE census in 2002; results not yet available MIKE census in 2002 + 500 in adjacent areas Comment NP – national park; GR – game reserve; CF – classified forest; MIKE – Monitoring Illegal Killing of Elephants Haut Sassandra CF Fresco CF Bossématié and other eastern CF Haut Bandama GR Azagny NP Taï NP 60 60 65 60 40 20 1500 1000–1500 1000 200 10–20 150 75 ± 25 50 160 1800 1000 800 100 Comoé NP Marahoué NP Population Site Table 1. Population estimates for elephants in Cote d’Ivoire Merz and Hoppe-Dominik 1991 Parren et al. 2002 Douglas-Hamilton et al. 1992 Kouadio in lit. 2004 Kobon in Blanc et al. 2003 Ouattara in lit. 2004 Parren et al. 2002 Merz and Hoppe-Dominik 1991 Caspary 1999 Roth et al. 1984 GTZ / FGU 1979 Douglas-Hamilton et al. 1992 Nandjui Awo in lit. 2004 Roth et al. 1984 Bouché 2002 Bouché 2002 Steinhauer-Burkhart 1984 GTZ/FGU 1979 GTZ/FGU 1979 personal observation personal observation Roth et al. 1984 Hoppe-Dominik 1989 Blanc et al. 2003 N. Hunter in lit. World Heritage Centre Roth et al. 1984 Merz 1982 Merz and Hoppe-Dominik 1991 Source Fischer Pachyderm No. 38 January–June 2005 Elephants in Cote d’Ivoire Bontioli Mali Burkina Faso Diefoula Logoniegue CF 10º N Warigue CF ODIENNE Boundiali CF FERKESSEDOUGOU Guinea Comoé Mont Gbande CF 9º N Foumbou CF Kinkene/Nyellepuo-Nzi CF Haut Bandama 8º N Mont Sangbe COTE D'IVOIRE Tiapleu Keregbo Haut Sassandra Goaso 7º N Mont Péko Scio Marahoué Abokoumekroa Beki-Bossematie Bia Djambamakrou Duekoué Tené Goin-Cavally ABENGOUROU YAMOUSSOUKRO N'Zo PFR 6º N Ghana Songan-TaminMabi-Yaya Davo Taï Sapo Grebo Go-Bodienou Okromodou ABIDJAN Niegré Ankasa Liberia Haut Dodo/Grah/Hana CF Fresco SAN PEDRO 5º N Azagny Barrobo HARPER 8º W Legend 7º W Elephant range international boundary known towns possible rivers and lakes sighting/sign 6º W 5º W 4º W 0 protected areas input zones 62.5 3º W 125 250 km Source: African Elephant Database Digital Chart of the World CF – classified forest PFR – partial faunal reserve Figure 1. Some protected areas and elephant distribution in Cote d’Ivoire. Pachyderm No. 38 January–June 2005 67 Fischer desperate situation of elephants in parts of West Africa and send out a plea for conservation action for elephant populations in West Africa that are still healthy or at least exist in viable population densities. Materials and methods I collected anecdotal information on the elephant population in Comoé National Park during my scientific work there between 1993 and the outbreak of civil war in September 2002. Additionally I visited other elephant sites in Cote d’Ivoire during that period—Taï NP, Abokouamékro Game Reserve, Marahoué NP, the Bossématié forest reserve system—and talked to researchers who conducted scientific or conservation work in these and other areas. Data collection did not follow a standardized protocol but was rather informal. Information on present and former elephant densities in Cote d’Ivoire was derived from the literature. Results Elephants outside protected areas Elephant populations mentioned by Roth and Douglas-Hamilton (1991) that existed in Bouaké and Dimbroko Districts as well as around Odienné and Korhogo have been extinct for many years, as are elephants close to Abidjan (Roth et al. 1984; pers. obs.). During my stays in Cote d’Ivoire (about 60 months between 1993 and 2004) hardly anyone mentioned the presence of elephants outside protected areas, and when mentioned, the elephants were said to be adjacent to protected areas. I assume that it is unlikely that any elephant population still exists outside protected areas, although it cannot be ruled out that the populations mentioned here sometimes leave protected sites. However, I assume these incidents are rare. Elephants within protected areas COMOÉ NATIONAL PARK Comoé NP, included in the World Heritage Site’s Danger List in 2003, is West Africa’s largest savanna park. The predominant vegetation type (wooded savanna with scattered forest islands and gallery forest), the annual rainfall (1000–1500 mm) and the permanent water supply of its major rivers make this park high-quality elephant habitat. There is no hu- 68 man encroachment, no livestock keeping, no illegal logging and no plantations within the 11,500 km2 of the park. However, poaching of all larger mammals is severe and has had catastrophic effects on its mammal populations (Fischer and Linsenmair 2001; Fischer et al. 2002). Elephants must have been very common both before and after the park was established. Local assistants told me that elephants came to the village frequently and drank from the well during the dry season in the late 1960s and early 1970s. There were no incidents of agricultural damage caused by elephants between 1993 and 2004 and hence no legal killing of problem animals. Elephant poaching was however common during that time. Elephants targeted included juveniles with very small tusks, hunted for ivory and meat. Three elephant carcasses (fig. 2) and one of a wounded calf (fig. 3) were found between 1993 and 1995, all animals that had obviously escaped poachers but died of severe wounds later. The real number of elephants killed by poachers during this time was probably much higher but could not be assessed accurately. Due to poaching, elephants were very shy and fled as soon as they became aware of human presence. Nevertheless, sightings of up to 120 elephants were made as late as 1997 (K.E. Linsenmair, pers. comm.) and elephant tracks and traces frequently seen (pers. obs.). From 1998 direct observation of elephants was extremely rare with only reported sightings after 1999, made by villagers in the southern part of the park close to the village of Gorowi. These sightings consisted of 11–15 animals, and probably always the same elephant group. Tracks and traces were hardly ever seen in the study area of approximately 200 km2 by researchers from the University of Würzburg after 1999. Even at wallows, salt licks and drinking sites in the gallery forest, which had been the major attraction points of elephants earlier (pers. obs.) no signs of elephants were seen here after 2000. The seasonal movements of elephants that are supposed to have occurred earlier (GTZ/FGU 1979) could not be determined by Steinhauer-Burkhart (1984) and definitely had come to a complete stop afterwards. MARAHOUÉ NATIONAL PARK Marahoué NP (1010 km2) is located in the transition zone between savanna and evergreen forest. No recent Pachyderm No. 38 January–June 2005 Elephants in Cote d’Ivoire Figure 2. Elephant carcass in Comoé National Park, Cote d’Ivoire, in 1996. elephant counts were made before the 2002 MIKE (Monitoring Illegal Killing of Elephants) counts, but during my visit to the park in 1997 local people reported several sightings of elephants, and tracks and traces were easily found. Large parts of the park were taken over by agriculturists with violent encounters between game wardens and farmers in 2001. Increase in human–elephant conflict is likely, in which case the elephants are sure to lose. Lack of funds made it impossible to manage the park after civil war broke out in 2002. The elephant numbers that Roth et al. gave in 1984 were significantly more than in the count shortly after that by Hoppe-Dominik (1989), and by 1991 the numbers were even fewer (Blanc et al. 2003). A MIKE team counted the remaining elephants in the park in July 2002 and estimated the population to be 160 in July 2002 with a confidence interval of 149–177 (N. Hunter 2005 in lit.). TAÏ NATIONAL PARK Taï NP is the last protected Haut-Guinean forest block in Cote d’Ivoire, covering 3300 km2 plus a 200-km2 buffer zone. It is contiguous to the N’Zo faunal reserve (730 km2). Elephant numbers decreased steeply in Taï NP despite the large amount of foreign aid pro- Pachyderm No. 38 January–June 2005 vided for well over a decade by different non-governmental organizations and government sources (Blanc et al. 2003). Logging, poaching, farming and illegal gold mining threaten the future entity of the park (World Heritage Centre 1989; pers. obs.). A big problem is the influx of refugees from Liberia and the back-and-forth movements since 2002 of Ivorian refugees, caused by civil war in Cote d’Ivoire. According to Roth et al. (1984) elephants in Taï NP were hunted mainly for their ivory but hunting for meat can also be assumed. Elephant poaching declined or even stopped altogether for several years, but the situation deteriorated recently given the desperate situation of the many refugees in the area. A MIKE team conducted an elephant survey in Taï NP in 2002 and results will be available soon. AZAGNY NATIONAL PARK Azagny NP in southern Cote d’Ivoire covers 190 km2. The area was historically difficult to reach but it might also have contained only relatively small areas of suitable elephant habitat. A. Nandjui (2005 in lit.) conducted an elephant survey in the park in 2003 and estimated about 65 elephants lived in the park. 69 Fischer BANCO NATIONAL PARK HAUT BANDAMA GAME RESERVE Close to the city of Abidjan is Banco NP, which measures 30 km2. Elephants are supposed to have been extinct here for almost a century (GTZ/FGU 1979) but the two to nine elephants killed close to Abidjan every year from 1975 to 1980 (Roth et al. 1984) might well have come from this park and adjacent areas. Today elephants are definitely extinct in the park and its surroundings. The numbers given by Roth et al. (1984) and Bouché (2002) show a steady decrease of elephants in the Haut Bandama reserve. Although there have been no counts recently it is unlikely that the number of elephants has increased significantly since the last estimate in 2002. BOSSÉMATIÉ FOREST SYSTEM Figure 3. A juvenile elephant killed by poachers in the southern Comoé National Park, Cote d’Ivoire, in 1993. 70 Close to the city of Abengourou, 1 to 22 problem elephants were killed annually between 1975 and 1980. An additional 30–60 elephants were poached in that area every year during the same period (local former hunters, pers. comm. 2004) leading to a steep decline in the elephant population in that zone. Today exact numbers do not exist, but the reserve forests near Abengourou host fewer than 60 elephants. No incidents of poaching have been observed during the last 10 years, but direct observations during standardized monitoring procedures became increasingly rare after 1999 (Goor 2004). Elephants were still present in the forest reserves of Bossématié, Songan, Tamin, Mabi and Yaya in late 2004 (pers. obs.). There were no reports of crop-raiding elephants (Parren et al. 2002) and Theuerkauf et al. (2001) showed that elephants even avoided villages and plantations. To ensure the long-term survival of this population of forest elephants Parren et al. (2002) suggested creating corridors to connect elephant populations in Cote d’Ivoire with suitable habitat in Ghana to enhance interbreeding of the subpopulations living on both sides of the border. The remaining forest elephant habitat in the Bossématié forest system has been under great risk ever since a long-term develop- Pachyderm No. 38 January–June 2005 Elephants in Cote d’Ivoire ment project financed by German aid agencies terminated in 2004. The lack of national funding and the demand for tropical timber is likely to lead to intensive logging in the area to generate income in the dwindling economy of Cote d’Ivoire. Long-term survival of the elephants is hence threatened by the potential destruction of their habitat in the near future. HAUT SASSANDRA FOREST RESERVE The elephant population in Haut Sassandra Forest Reserve has been stable during the last seven years. Updated information is not available for sites that hosted elephants earlier and that were listed in 2003 by Blanc et al. Discussion Data quality and knowledge gaps Data quality among sites differed substantially. The best data came from the MIKE census in 2002 for Marahoué NP and from A. Nandjui for Azagny NP in 2003 (A. Nandjui 2005 in lit.). Some good recent guesses are probably those for Comoé NP. This emphasizes the need to perform comprehensive elephant counts for all sites that still contain or recently contained elephant populations. We need data on population densities, sex and age ratios as well as reproduction rates to complete the picture of the current situation of the Ivorian elephants and properly evaluate the status of each population. Elephant research between 1950 and the late 1990s showed a strong bias for work in eastern Africa, followed by that in southern and Central Africa with only 5.2% of the published studies conducted in West Africa. An overall decrease of elephant publications began in the late 1980s (Bossen 1998). This again emphasizes the importance of elephant studies in West Africa. Estimation of elephant population densities and trends Elephant numbers in Cote d’Ivoire have decreased at alarming rates since the first population surveys were conducted, and warnings that this species might be lost were put forward in 1984. Roth and DouglasHamilton (1991) referred to Comoé NP as one of the most important elephant ranges in West Africa and called for its improved management and protection as vital for the long-term survival of savanna elephants Pachyderm No. 38 January–June 2005 in the region. Despite this strong statement made by Roth et al. as early as 1984 no efforts were made to protect the park and its elephants, leading to a further decrease with only 10–20 animals estimated to be remaining in the park today. The only population that seems to have increased is in Marahoué NP. However, I believe that this is due rather to the inaccuracy of earlier counts than to a substantial increase in numbers. Elephant poaching in the park was rare in the late 1990s (pers. obs.), which might have allowed for a certain regeneration of the population, but human encroachment was severe, with land being used in ways that are not compatible with elephant presence. Although the size of a viable elephant population is unknown (IUCN 2003a), Cote d’Ivoire might not host even one population that would survive for long or have the potential to increase significantly. While Sukumar (1993) suggested that a viable population of Asian elephants had to contain 100–200 animals, depending on its structure, numbers given by Vucetich and Waite (1998), independent of the species in question, are much higher. Despite this uncertainty, elephant populations in Cote d’Ivoire may be too small and too isolated from each other to ensure their regeneration. They are additionally threatened by a complete lack of recolonization events and by synchronous trends towards decreasing numbers in all subpopulations, which has been stated by Barnes (1999) as a major negative effect for recovery of elephant populations in West Africa. Even though the size of the remaining populations is largely unknown, the Marahoué NP population is probably the only one containing more than 100 individuals. Management implications Small elephant populations in dense vegetation are almost impossible to count precisely (Barnes 2002) which makes trend detection in all Ivorian populations largely impossible. Dung counts and genetic analysis of faeces are the most promising techniques to apply (Barnes 2001). Even if effective conservation measures are taken immediately, which is doubtful due to the political situation in the country at the moment, elephant numbers might be too small and the population structure too imbalanced to allow for population regeneration. With the possibility that there are two African elephant species (Roca et al. 2001), conservation fo- 71 Fischer cus should be laid on the forest elephant Loxodonta cyclotis. Although elephants are probably not essential for the future existence of the remaining West African forests, they play an important ecological role in the succession dynamics of these forests (Hawthorne and Parren 2000). In general, little is known about their ecological role in West Africa, where they function as ‘landscape architects’ and seed dispersers of economically and environmentally important plants. Given this lack of knowledge and taking into account the fact that many tree species are long lived, the ecological disaster resulting from their dying out might not emerge until well after their extinction. The situation for savanna elephants looks even worse. While restoring and improving habitat through different management schemes as proposed by the IUCN Conservation Strategy (2003a) is important for many areas, doing so is of minor concern for the savanna habitat of the elephant (mainly Comoé NP). The park’s single problem was and is poaching, with the habitat not only for elephants but for all large mammals being still intact. Because it is large, Comoé NP could host a large and viable elephant population provided it were truly efficiently protected and managed. The human footprint map of Cote d’Ivoire (http:/ /wcs-old.atlasworks.com/media/file/hf_IvoryCoast1. pdf) shows that human impact in the north-east of the country (except for poaching) is moderate to low. Since there is so little genetic difference among African savanna elephants (Roca et al. 2001) and elephant habitat is still intact, Comoé NP could one day become a site for reintroducing elephants from other parts of Africa. Management and protection of the park would need to be improved considerably and genetic distinctiveness taken into account (see Eggert et al. 2002 for details). Eggert et al. (2002) suggested the existence of three elephant species in Africa with western savanna elephants forming a distinctive group. If this is true, protecting and conserving the remaining West African savanna elephants becomes a much higher priority. Transborder protected areas between Cote d’Ivoire and neighbouring countries were suggested by Soulemane (2002) to reduce human–elephant conflict and by Parren et al. (2002) to improve interbreeding of subpopulations with neighbouring Ghana. Such cooperative approaches are important to achieve maximum benefit of pooled resources and secure elephant ranges that span international borders (IUCN 72 2003a). Additionally they might be the only measure that can prevent elephant extinction in Cote d’Ivoire since local elephant populations are too small to ensure their own long-term survival without input from other areas. Transborder protected areas can also ensure elephant survival during times of civil strife, as happened in the 1990s when 300 elephants moved from Togo to adjacent areas in Benin and Burkina Faso (Ph. Bouché, pers. comm.). A detailed action plan for two transborder protected areas in southern Cote d’Ivoire with Liberia and with Ghana has been proposed by IUCN (2003b) but not yet implemented, and similar proposals exist for a transborder protected area between Comoé NP and Forêt Classée et Réserve Partielle de la Faune de la Comoé Léraba in Burkina Faso (Oliver Hamerlyck, pers. comm.). The establishment of bio-corridors with areas in Burkina Faso and Ghana should be considered, not only to enable elephants to repopulate Comoé NP but also to connect metapopulations of other large mammals. A national elephant conservation strategy for Cote d’Ivoire was discussed during a workshop in Abidjan in December 2003, but the results are not yet freely available. Legal situation Although hunting female and young elephants became illegal in 1965 and elephant hunting was completely banned in 1974 in Cote d’Ivoire (Roth et al. 1984) problem elephants could still be killed. Roth et al. (1984) claimed that around 90 elephants were legally killed every year between 1975 and 1980 despite the very low frequency of crop damage they caused. Even though agricultural damage by elephants was little as they largely avoided cacao and coffee plantations (Theuerkauf et al. 2001), farmers still complained about rare incidents of crop damage by problem elephants to get hunting permits (Soulemane 2002). Despite the legal ban on domestic ivory trade in 1997, raw and worked ivory products could be found throughout the country even much later (TRAFFIC 2003; pers. obs.). Cote d’Ivoire not only exported ivory but was a major importer of elephant tusks until 1980 (Roth and Douglas-Hamilton 1991). Pachyderm No. 38 January–June 2005 Elephants in Cote d’Ivoire Underlying forces The human population in Cote d’Ivoire increased about sixfold between 1950 and 2004; during the same period almost 90% of the forest disappeared (Bryant et al. 1997). And during the entire period, elephant numbers decreased even faster than the available habitat shrank. While the increase in human population is assumed to have slowed down (from over 3 to 2.1%) due to the HIV/AIDS epidemic (CIA World Fact Book 2004) destruction of natural habitats is likely to increase due to heavy dependence of the local economy on the production of cocoa and coffee, most of which is produced on sensitive elephant habitat. Due to economic problems after the outbreak of civil war in 2002, nature conservation, which was weak and a low priority even before the civil strife, will probably be of even less importance in the near future. The negative effects that occurred in Comoé NP (Fischer 2004), despite its low human population pressure, will most likely be more severe in the southern parts of the country that contain higher human population densities. Outlook Elephants in Cote d’Ivoire occur more or less exclusively inside protected areas. However, a complete lack of management in the largest legally protected area—Comoé NP—and weak protection of other national parks in which elephants occur, means the survival of this species even inside protected areas in Cote d’Ivoire is doubtful. National parks like Marahoué and other sites such as the Bossématié forest system are threatened by human encroachment, increasing agriculture and in Bossématié even logging of the entire elephant habitat. Since the relatively large population (200 animals estimated in 1997) in Comoé NP collapsed later, elephants are most likely at the brink of extinction all over Cote d’Ivoire with little chance of survival due to ongoing habitat destruction (for forest populations), hunting (for savanna populations) and genetic effects such as inbreeding and sex imbalance for both forest and savanna populations. Conclusions Although the elephant is the national emblem of Cote d’Ivoire, with a powerful political party using it as its symbol and the national football team called Les Pachyderm No. 38 January–June 2005 Eléphants de Cote d’Ivoire, no effort has been made to protect the remaining elephants in the country at any time. Elephants are depicted in the logos of many Ivorian companies, and elephant sculptures and illustrations are found in plenty in handicraft markets, yet most Ivorians are little aware or concerned that elephants are about to vanish from their country. The loss will be realized only after the species becomes extinct. However, Cote d’Ivoire should not be blamed alone; despite their permanent presence in the country no major international NGO has actively worked to protect Ivorian elephants within the last decade despite early warnings from several different sources. Although pleas for action to save the elephants of Cote d’Ivoire were made as early as 1984 (Roth et al. 1984), no measures have been taken, resulting in the probable extinction of the species in one of its former strongholds in the near future. One lesson learned is that the time to act might be running out for other sites as well and that immediate action should be taken for those locations that contain viable elephant populations in West Africa. The political situation in Cote d’Ivoire went from stability to civil strive within a very short time, which resulted in even further deterioration of the conservation situation in the country and its major national park (Fischer 2004). This should be a warning example for other areas that might be politically stable at the moment and still have relatively high numbers of elephants. The last strongholds for savanna elephants in West Africa are the ‘W’–Arli–Pendjari–Oti Mandouri–Keran (WAPOK) ecosystem in Benin, Burkina Faso, Togo and Niger (Bouché et al. 2004a), the Nazinga Ranch in Burkina Faso (Bouché et al. 2004b) and to a lesser extent Mali’s Gourma area and Mole NP in Ghana (Blanc et al. 2003; Brice Sinsin, pers. comm.). The WAPOK ecosystem hosted 4600 elephants in 2003 with 71.5% of this population counted in Burkina Faso, 26.6% in Benin and 2% in Niger (Bouché et al. 2004a). According to Ph. Bouché (pers. comm.) the 940-km2 Nazinga Game Ranch contains the highest density of elephants in West Africa with an estimated 550 animals and an average annual increase of 3.8% between 1989 and 2003. The above-mentioned sites have to be of highest conservation priority, with the poor example of Cote d’Ivoire kept in mind. It is alarming to note that the elephant population in Mole NP decreased between 1993 and 2004 (Ph. Bouché, pers. comm.). Since future efforts will be concentrated on areas that have populations of more than 100 elephants 73 Fischer (IUCN 2003a), Cote d’Ivoire will no longer be a focus country for elephant conservation, with small populations like the Comoé elephants most likely facing extinction soon. To establish efficient conservation programmes for forest elephants Blake and Hedges (2004) asked for sound data collection to fill knowledge gaps and for continued monitoring of existing populations to ensure future existence. I support their request and ask that it be expanded to include West African savanna elephants. Intensive data collection on population structures and trends should be carried out before priorities are set and populations with the highest probability of long-term survival determined (IUCN 2003). However, data collection should go hand in hand with anti-poaching and other management measures as elephant populations can be depleted within a short time if left unattended (Merz 1982; pers. obs.). Any of these activities will probably have to focus on areas outside Cote d’Ivoire, most likely the WAPOK, Gourma and Mole protectedarea systems. Remaining elephant populations must be managed and protected across political and habitat borders, as has been suggested for the jaguar (Panthera onca) by Sanderson and colleagues (2002). I want to emphasize the importance of the remaining elephant populations in West Africa and plead for combined efforts to prevent these populations from undergoing the same fate as the elephants in Cote d’Ivoire. Elephants are charismatic animals, and their drastic decrease in numbers in West Africa should be used to attract more scientific and conservation activities to this geographical area. Acknowledgements I thank the Ministère de l’Agriculture et des Ressources Animals in Abidjan for a permit (1223/Miniagra/ CAB-1) to conduct research in Comoé NP. Thanks too to Holly Dublin for suggesting this publication and to Akoi Kouadio, Nandjui Awo and Nigel Hunter for sharing their data and information with me. I would like to thank Richard Barnes, Philippe Bouché and two anonymous referees for critical comments on an earlier version of this paper. References Barnes RFW. 1999. Is there a future for elephants in West Africa? Mammalian Review 29(3):175–199. 74 Barnes RFW. 2001. How reliable are dung counts for estimating elephant numbers? African Journal of Ecology 39:1–9. Barnes RFW. 2002. The problem of precision and trend detection posed by small elephant populations in West Africa. African Journal of Ecology 40:179–185. Blake S, Hedges S. 2004. Sinking the flagship: the case of forest elephants in Asia and Africa. Conservation Biology 18(5):1191–1202. Blanc JJ, Thouless CR, Hart JA, Dublin HT, Douglas-Hamilton I, Craig CG, Barnes RFW. 2003. African elephant status report 2002. IUCN/SSC African Elephant Specialist Group, Gland, Switzerland. Bossen B. 1998. Research on African elephants Loxodonta africana (Blumenbach, 1797): a bibliography. African Journal of Ecology 36:371–376. Bouché Ph. 2002. Elephant status and conservation in the Upper Bandama Game Reserve, Ivory Coast. Pachyderm 32:72–73. Bouché Ph., Lungren CG, Hien B, Omondi P. 2004a. Aerial total count of the ‘W’–Arli–Pendjari–Oti Mandouri– Keran (WAPOK) ecosystem in West Africa. http:// w w w. c i t e s . o r g / c o m m o n / p r o g / m i k e / s u r v e y / WAPOK_survey03.pdf Bouché Ph., Lungren CG, Hien B. 2004b. PONASI ecosystem survey. http://www.cites.org/common/prog/ mike/sub_reg/PONASI_survey.pdf Bryant D, Nielsen D, Tangley L. 1997. The last frontier forests. World Resources Institute,Washington, DC. Caspary HU. 1999. Wildife utilization in Cote d’Ivoire and West Africa: potentials and constraints for developmental cooperation. GTZ, Eschborn. CIA World Fact Book. 2004. http://www.cia.gov/cia/publications/factbook/geos/iv.html Eggert LS, Rasner CA, Woodruff DS. 2002.The evolution and phylogeology of the African elephant inferred from mitochondrial DNA sequence and microsatellite markers. Proceedings of the Royal Society of London, Series B 269:1993–2006. Fischer F. 2004. Status of the Comoé National Park, Cote d’Ivoire, and the effects of war. Parks 14(1):17–25. Fischer F, Linsenmair KE. 2001. Decreases in ungulate population densities. Examples from Comoé National Park, Ivory Coast. Biological Conservation 101(2):131–135. Fischer F, Gross M, Linsenmair KE. 2002. Updated list of the larger mammals of Comoé National Park, Ivory Coast. Mammalia 66(1):83–92. Goor W van. 2004. Assessing the status of mammals in the Forêt Classé de la Bossématié, Cote d’Ivoire with its influencing factors. BSc thesis. University of Larenstein. Pachyderm No. 38 January–June 2005 Elephants in Cote d’Ivoire GTZ/FGU. 1979. Gegenwärtiger Status der Como und TaïNationalparks sowie des Azagny-Reservats und Vorschläge zu deren Erhaltung und Entwicklung zur Förderung des Tourismus. PN 73.2085.6 Band II Teil 1 + 2. Gesellschaft für Technische Zusammenarbeit, Eschborn. Hawthorne WD, Parren MPE. 2000. How important are forest elephants to the survival of woody plant species in Upper Guinean forests? Journal of Tropical Ecology 16:133–150. Hoppe-Dominik B. 1989. Premier recensement des grandes mammifères dans le Parc National de la Marahoué en Cote d’Ivoire. African Journal of Zoology 103:21–27. IUCN. 2003a. Strategy for the conservation of West African elephants. IUCN, Gland, Switzerland. IUCN. 2003b. Action plan for the management of transfrontier elephant conservation corridors in West Africa. IUCN, Gland, Switzerland. Merz G. 1982. Untersuchungen über Lebensraum und Verhalten des afrikanischen Waldelefanten im Taï Nationalpark der Republik Elfenbeinküste unter dem Einfluss der regionalen Entwicklung. Dissertation, University of Heidelberg. Merz G, Hoppe-Dominik B. 1989. Distribution and status of the forest elephant in the Ivory Coast, West Africa. Pachyderm 14:22–24. Parker ISC. 1989. Elephant decline, part 1: downward trends in African elephant distribution and numbers. International Journal of Environmental Studies 34:287–305. Parren MPE, de Leede BM, Bongers F. 2002. A proposal for a transnational forest network area for elephants in Cote d’Ivoire and Ghana. Oryx 36(3):249–256. Roca AL, Georgiadis N, Pecon-Slattery J, O’Brian SJ. 2001. Pachyderm No. 38 January–June 2005 Genetic evidence of two species of elephant in Africa. Science 293(5534):1473–1477. Roth HH, Merz G, Steinhauer B. 1984. Répartition et statut des grandes espèces de mammifères en Cote d’Ivoire. Mammalia 48(2):207–226. Roth HH, Douglas-Hamilton I. 1991. Distribution and status of elephants in West Africa. 1. Mammalia 55(4): 489–527. Sanderson EW, Redford KH, Chetkiewicz Ch-L, Medellin RA, Rabinowitz AR, Robinson JG, Taber AB. 2002. Planning to save a species: the jaguar as a model. Conservation Biology 16(1):58–72. Soulemane O. 2002. Conflits homme–éléphant autour de la Forêt classée du Haut-Sassandra, Cote d’Ivoire. Pachyderm 32:12–21. Steinauer-Burkhart B. 1984. Untersuchung zur Ökologie von Großsäugern anhand von Flugzählungen im Comoé National Park. Dissertation, University of Würzburg. Sukumar R. 1993. Minimum viable populations for elephant conservation. Gajah 11:48–52. Theuerkauf J, Ellenberg H, Waitkuwait WE, Mhlenberg M. 2001. Forest elephant distribution and habitat use in the Bossématié Forest Reserve, Ivory Coast. Pachyderm 30:37–43. TRAFFIC. 2003. More ivory than elephants: domestic ivory markets in three West African countries. TRAFFIC Online Report Series No 8. Vucetich JA, Waite Th. A. 1998. Number of censuses required for demographic estimation of effective population size. Conservation Biology 12(5):1023–1030. World Heritage Centre. 1989. http://www.wcmc.org.uk/ protected_areas/data/wh/tai.html 75 Noupa Analyse biométrie des pointes d’éléphants saisies dans le cadre de la lutte antibraconnage par les services de la conservation dans le massif du Sud-est Cameroun Paul Noupa, Consultant/Senior Conservation Biologist IUCN Regional Office for Central Africa, PO Box 5506, Yaoundé, Cameroon; email : [email protected] Résumé La biométrie aide à l’analyse des variations biologiques à l’intérieur d’un groupe déterminé. Dans le cadre des populations d’éléphant de la forêt du Sud-est Cameroun, elle contribuerait à l’amélioration de la connaissance sur la dynamique de sa population, ce qui aiderait à l’aménagement de l’espèce. Pour le faire, la présente étude s’est occupée à relever les mensurations sur les pièces dures de sujets morts (abattus par braconnage) en vue d’établir une éventuelle relation dans la variabilité de divers paramètres biologiques (âge, poids, sexe, …). A terme, l’établissement de cette relation entre les mesures des pièces dures et les paramètres biologiques individuels permettent une caractérisation de la population en terme d’identification du pourcentage des sujets jeunes, sub-adultes, adultes et vieux). La mensuration de la présente étude a porté sur 62 pointes d’ivoire et révèle que 2 éléphants sur 3 abattus dans le Sud-est Cameroun sont de jeunes sujets. En effet, la moyenne de poids des pointes d’ivoire pour les animaux abattus par les braconniers est de 4,44 kg contre une moyenne dans la région de 15 kg. Ceci est un indicateur du type de prélèvement fortement préjudiciable à l’espèce que font les braconniers sur l’éléphant. Cette analyse a aussi permis de constater qu’un dimorphisme sexuel s’installe chez les jeunes éléphants lorsque les pointes d’ivoire atteignent une longueur de 75 cm. Bien que faite sur un échantillon réduit, cette étude pose les jalons de la confection des abaques pour les populations d’éléphant dans le région du Sud-est si elle est complétée par une collecte patiente des données qui mette à contribution les guides chasse professionnels travaillant dans la région. Abstract Biometrics analyses morphological variation within a given group. Within the context of forest elephant populations in south-eastern Cameroon, biometrics can contribute to improving knowledge on population dynamics, thus aiding in management of the species. This study uses measurements of body parts taken from poached elephant specimens to investigate the relationships between biometric measurements and a number of biological parameters (age, weight, sex...). The characterization of such relationships permits the investigation of population structure in terms of juveniles, subadults, adults and old adults. Measurements of 62 tusks revealed that 2 out every 3 elephants poached in south-eastern Cameroon are young individuals. The average tusk weight among the poached elephants examined was 4.44 kg, against an average of 15 kg for the area. This provides an indicator of the type of offtake prevalent in the area and its potential impact on the species. This analysis also provides evidence that sexual dimorphism becomes apparent in young elephants when the tusk length reaches 75 cm. Although conducted on a small sample, this study sets the stage for detailed knowledge of elephant population structures in south-eastern Cameroon. This effort should be supplemented with additional data collected by professional hunting guides working in the area. 76 Pachyderm No. 38 January–June 2005 Analyse biométrie des pointes d’élephant saisies au Cameroun consiste à la création et l’aménagement de trois parcs nationaux—Lobeke, Boumba-Bek et Nki (8000 km2)—et leur zone périphérique pour la conservation de la biodiversité. Dans le cadre de la surveillance dans le site, le programme conduit des opérations de démantèlement des campements de braconniers qui ont conduit à la saisie d’un stock de pointes d’ivoires qui ont été utilisés dans la présente étude. Introduction La forêt du sud-est appartient à l’écosystème de la forêt dense humide de basse altitude du nord-ouest du bassin du Congo. Elle représente un bloc forestier d’habitat peu perturbé de l’aire de répartition de l’éléphant de forêt (Loxodonta africana cyclotis) au Cameroun (fig. 1). C’est une forêt reconnue pour sa richesse en essences commerciales et qui a favorisé le développement d’une intense activité d’exploitation depuis environs trois décennies avec environ 29 essences forestières de haute valeur commerciale jusqu’ici prélevées (MINEF 2001). Elle est aussi exceptionnellement riche en espèces fauniques parmi les plus charismatiques et les plus menacées à savoir le chimpanzé, le gorille, le drill et l’éléphant. La riche biodiversité de la région a été internationalement reconnue et sa préservation est connue comme nécessitant l’effort conjugué des gouvernements, des ONG et autres donateurs. Dans ce massif, des initiatives de conservation sont engagées par le programme de gestion durable de la biodiversité du sud-est qui a bénéficié de l’appui financier du GEF/Banque Mondiale, de la GTZ et des fonds du Pays-Bas à travers le projet du WWF dans la région. Cette initiative de conservation dans le site Matériel et méthode Chez les espèces animales, les pièces dures constituent des parties du corps qui résistent longtemps après la mort du sujet. Ces parties dures sont diverses et variées, mais généralement, en font partie les os et les éléments dentaires. Par ailleurs, certaines de ces pièces s’installent dans la vie de l’individu après la naissance. Mais généralement, ces pièces subissent un accroissement de taille, de grosseur, de hauteur, de poids et même de forme avec le développement physique du sujet. Dans certains cas, ces pièces subissent des transformations marquées qui permettent de déterminer les changements d’année, ce qui a permis qu’on les utilise pour déterminer les âges des sujets. Dans le cas de cette étude, les pointes d’ivoires ont été utilisées pour essayer de caractériser Gari-Gombo N Yokadouma Mboy II Moampack Ngato Cameroun Bangué Massea Malea ancien PN de Libongo BOUMBA BEK Ngoila Koumela PN de NKI Mambélé PN de LOBEKE Nguilili Légende Ndongo Zone communautaire de PN de Lobeke Moloundou PK 27 PK 14 Djembé Kika routes PN Parc national 40 0 40 80 km Socambo Pachyderm No. 38 January–June 2005 77 Noupa Résultats Répartition des saisies par station Entre 1998 et 2000, 62 pointes d’ivoire ont été saisies dans le sud-est, dans quatre stations et environ 80 % provenaient de la station de Kika qui est la pointe sudest du Cameroun caracterisée par la présence des parcs nationax de Boumba Bek, Nki et Lobeke (fig. 1). Cette région constitue naturellement un refuge pour la population d’éléphants refoulée par les perturbations des activités économiques et sociales des populations Une répartition des pointes saisies par catégorie de poids et par station est faite dans la figure 2. Selon la réglementation en vigueur sur la faune au Cameroun, les sujets ayant des pointes de moins de 5 kg sont jeunes et par conséquent intégralement protégés. Or il ressort des mensurations que 74 % des 78 saisies soit 45 pointes sur 62 (fig. 3) ont un poids inférieur à 5 kg et, au sens de la disposition réglementaire, proviennent d’une catégorie de population intégralement protégée. L’abattage par les braconnier se faisant au hasard sans aucun souci de sélection de sujet à abattre, il ressort de cette information que les populations dont sont issues ces pointes sont majoritairement jeunes et ceci concerne 2 éléphants sur 3 dans l’échantillon. Cette variation est aussi perçue à travers la moyenne de poids des pointes qui est de 4,44 kg dans l’échantillon d’étude contre 15 kg en moyenne pour les vieux sujets dans la région et où le plus vieux sujet jamais enregistré a eu une pointe de 35 kg (record national d’éléphant de forêt). Les braconniers dans leur destruction aveugle, abattent les sujets qui sont au tiers de leur longévité. 90 80 70 Pourcentage 60 50 40 30 20 10 0 Kika Ngato Ancien Koumela Mboy II Station Figure 2. Répartition des saisies des pointes d’ivoire dans la région du sud-est Cameroun entre 1998 et 2000. (Source : Action du Programme de conservation et de gestion de la biodiversité dans le Sud-est) Nombre les sujets porteurs et tenter de comprendre les caractéristiques de la population dont ils sont issus. Les pièces durs pourraient permettre des mensurations qui, complétées par des données prélevées sur les sujets entiers nouvellement abattus dans la région, peuvent aider à produire des abaques relationnelles qui caractérisent la population. A cause de l’insuffisance des informations recueillies sur l’échantillon de pointe qui était disponible pour cette étude, elle ne prétend pas établir la variation biologique au sein de la population de laquelle sont issus les sujets échantillonnés. Ainsi, l’étude ne peut pas répondre à la préoccupation de la relation entre les mensurations prélevées sur les pointes d’ivoire et les informations sur l’âge, le poids, la taille, du sujet, toutes choses nécessaires pour confectionner des abaques caractéristiques des populations, mais elle permet de valoriser les informations souvent difficiles d’accès dans le contexte de l’aménagement des éléphants de forêt dans cette région qui est aussi un foyer de braconnage des éléphants. L’étude a porté sur la mensuration de 62 pointes d’ivoire issues des saisies lors des activités de lutte anti-braconnage menées dans le site entre 1998 et 2000. Ces pointes d’ivoire ont été mesurées et pesées et les informations suivantes ont été rassemblées sur chaque pointe d’ivoire : origine, hauteur de la pointe, courbure externe, diamètre au centre, aplatissement au centre, poids en kg et aspect physique. Ces pointes ont été saisies dans quatre stations : Kika, Ngato Ancien, Koumela et Mboy II (fig. 1). Les données de mensuration ont été encodées avec le logiciel Excel. 50 45 40 35 30 25 20 15 10 5 0 Pointe de moins de 5 kg Pointe de plus de 5 kg Kika Ngato Koumela Mboy II Ancien Station de saisie T Total Figure 3. Répartition des pointes d’ivoire saisies par catégorie de poids et par station. Pachyderm No. 38 January–June 2005 Analyse biométrie des pointes d’élephant saisies au Cameroun de forêt à prendre des mesures pour la protection de la jeune population d’éléphants dans le massif du sudest Cameroun. La population d’élé-phants de cette région est soumise à la pression de la chasse sportive, du braconnage et de la dégradation de l’habitat. Il y a un besoin d’aména-gement et de stabilisation de cette population au risque que le braconnage rampant continue à déprécier la population de cette espèce avec à terme la dégradation du potentiel génétique par une réduction de la population à un seuil non viable. Ceci interpelle principale-ment les initiatives en cours dans la région pour l’éléphant, notamment : Le programme de conservation et de gestion de la biodiversité dans le sud-est Cameroun, les programmes spécifiques tels que MIKE/CITES (Monitoring Illegal Killing of Elephant), le Groupe de travail sous-régional sur la viande de brousse, le Groupe Spécialiste des Eléphants d’Afrique (GSEAf). L’étude a permis de distinguer deux phases dans le processus de développement des pointes à savoir : i) Une première phase qui englobe les pointes de longueur comprise entre 32 et 75 cm et de poids inférieur ou égale à 2 kg. Dans cette phase il existe une relation proportionnée entre le poids de la pointe d’ivoire et sa longueur indépendamment de la station. Cette phase concerne la tranche des individus « bébés ». Dans cette phase, il y a une relation linéaire entre la croissance en longueur et le poids de la pointe d’ivoire suivant la relation y = 0,0536x + 0,3766 Croissance des sujets dans la population échantillonnée La figure 4 représente la relation de poids à la longueur de chaque pointe pour les saisie des stations de Kika, Ngato Ancien, Koumela et Mboy II. Il ressort de ce graphique indépendemment des stations, une distincton entre deux catégories de populations d’éléphants : les populations jeunes (poids des pointes inférieur à 5 kg et de longeur au plus égale à 85 cm) qui représentent 70 % de l’échantillon étudié et les populations plus âgées (poids des pointes supérieure à 5 kg et longeur comprise ente 98 et 130 cm) qui représentent 30 % d l’échantillon étudié). Relation biométriques La figure 5 représente la relation biométrique l’échantillon d’étude. Elle donne les tendance d’évolution de la hauteur, de la courbure externe du diamètre médian en fonction du poids de la pièce. Discussion Pachyderm No. 38 January–June 2005 137 130 126 117 99,5 96,8 89 92,5 84 75 72 62 65,5 56,5 53,8 48,7 43,9 Poids (kg) Il ressort des résultats de cette étude que la station de Kika qui se situe dans la zone où on trouve encore une forte densité d’éléphants du fait de la presence des trois parcs nationaux de Lobéké, Boumba-Bek et Nki est aussi la région où on rencontre la plus fort activitée de braconnage. En effet, les autres zones qui connaissent une forte concen18 tration des structures et infrastructures humaines durables ou 16 permanentes (villages, champs) 14 et qui sont par conséquent fortement pertubées abritent de 12 faibles populations d’éléphants 10 et les saisies y ont été moindres 8 avec moins de 5 % de saisies à Koumela et Mboy II. 6 L’étude a aussi montré que 4 l’activité de braconnage dans la région du sud-est Cameroun 2 réduit à moins du tiers 0 l’espérance de vie de la populations d’éléphants. Cette situation doit interpeller Kika Ngato Ancien Koumela l’administration du Cameroun Longueur (cm) et tous ceux qui sont intéressés par la protection des éléphants Figure 4. Relation poids–longueur par station. Mboy ll 79 Noupa 180 y = -1,9309x + 148,08 R2 = 0,9112 160 Hauteur en cm Courbure externe en cm Diamètre au centre en cm Courbe de tendance de la courbure externe Coubre de tendance de la hauteur de la pointe 140 120 Valeur 100 80 y = -1,6041x + 128,26 R2 = 0,9008 60 40 0,38 0,55 0,71 0,86 1,1 1,2 1,35 1,5 1,7 1,85 1,9 2,6 3,2 4,3 5 7 8 10,5 12 14,5 0 17 20 Poids de la pointe (kg) Figure 4. Relation biométrique des pointes d’ivoire. où y représente la longeur de la pointe et x représente son poids en kg. La coefficient de corrélation de cette éqution de tendance est fort de R2= 0,81. Dans cette phase de croissance, il n’y aurait pas d’influence du sexe dans le développement somatique de l’individu. ii) Une deuxième phase survient où on note une forte variation entre les poids des pointes d’ivoire de même longueur et même une variabilité liée à la station. Il s’agirait de la phase d’installation du dimorphisme sexuel des sujets où la croissance des sujets mâles diffère de celle des sujets femelles. Il s’agirait aussi de la phase où l’influence de l’environnement a un effet sur le développement somatique des individus. Cette phase s’accompagnerait d’une différentiation dans la croissance des pièces dures. L’équation linéaire décrivant cette phase indépendamment de la station est y = 0,424x + 2,8664 avec un coefficient de corrélation R2=0,5525. L’influence de la différenciation sexuelle sur le développement des pointes d’ivoire des éléphants de la région du sud-est se met en place lorsque les pointes pèsent encore environ 4 kg. Malheuresement, les sujets sont beaucoup plus vite victimes de braconnage qu’on ne put apprécier l’évolution 80 de ce dimorphisme sur des sujets plus vieux. (Le record national des pointes d’ivoire pour les éléphants de forêt du Cameroun est un sujet dont une pointe a pesé 36 kg avec 2,40 m de longueur, abattu dans la région de Mintom). Relations biométriques Les résutats bimétriques de cette étude suggèrent qu’il existerait une corrélation positive entre la hauteur de la pointe, sa courbure externe et son poids. La relation entre la courbure externe de la pointe et le poids pour l’échantillon de saisie est donnée par l’équation y = –1,9309x + 148,08 où y représente la valeur de la courbure et x le poids avec un coefficient de régression R2 = 0,9112. De même, la relation entre la hauteur de la pointe et le poids pour l’échantillon de saisie est donnée par l’équation y = –1,6041x + 128,26 avec un coefficient de régression R2 = 0,9008. Les deux paramètres biométriques « Courbure externe et Hauteur de la pointe » sont corrélés au poids de la pointe à plus de 90 % de probabilité. Par contre la mesure du diamètre au centre de la pointe ne donne pas une relation précise avec le poids de la pointe. Pachyderm No. 38 January–June 2005 Analyse biométrie des pointes d’élephant saisies au Cameroun Même si la tendance générale est celle d’une corrélation positive entre le poids de la pointe et son diamètre au centre, il reste que des observations donnent des pointes de moindre poids présentant de grande diamètres et inversement. En réalisant ainsi qu’il y a une relation entre le poids et les paramètres biométrique des pointes d’ivoire, on est en présence d’un outil d’aménagement des populations d’éléphants de forêt. En effet, pratiquement, il est très difficile de prélever les paramètres caractéristiques (poids, âge, longueur, hauteur au garrot…) sur les sujets vivants dans la nature. Par contre, on rencontre très régulièrement les pièces dures (pointe d’ivoire) d’éléphant en circulation. La mise en relation des paramètres des pièces dures (poids, hauteur, courbure externe…) avec les relevés des paramètres sur un échantillon des sujets vivants donne une abaque relationnelle qui permet de savoir pour une population donnée, lorsqu’on est en présence de pièce dure, quel a été l’âge et le poids de ce sujet. Un tel outil serait d’une importance capitale pour la gestion de la chasse sportive et l’aménagement des populations d’éléphant de forêt. Conclusion De l’analyse des saisies des pointes d’ivoire dans le sud-est entre 1998 et 2000, il ressort que la population des éléphants de la région sont fortement composée d’individus jeunes ayant les pointes inférieures à 5 kg. Les présentes mensurations sur les pointes saisies dans le sud-est du Cameroun ont permis de vérifier qu’il existe une corrélation positive entre la hauteur des pointes d’ivoire, leur courbure externe et leur poids. Les données disponibles ne permettent pas de lier ces informations à l’âge de l’individu, son poids et les autres paramètres quantitatifs (envergure) caractéristiques des individus qui, mis ensemble permettent aussi de caractériser les populations. Toutefois, il est nécessaire de réaliser une étude biométrique des pointes d’ivoire pour avoir des informations sur l’âge de ces populations et de construire pour la région un abaque qui sera un outil important pour l’aménagement de cette espèce et de son habitat. La mise en place d’un protocole en vue de la préparation d’un tel outil incombe certes au scientifique, mais elle requiert la collaboration et la discipline des services à charge de la gestion de la faune et des guideschasse travaillant dans les zones affermées. Pachyderm No. 38 January–June 2005 Dédicace Dr Paul Robinson Ngnengue, décédé le 21 août 2001, pour sa contribution au développement du Projet de conservation du WWF dans le sud-est Cameroun et notamment la mise sur pied d’un système de suivi écologique. A sa mort, Robinson était le responsable du suivi-écologique du Jengi Southeast Forest Project. Bibliographie de référence sur le site d’étude Barnes RFW, Jensen KL. 1987. How to count elephants in forests. IUCN African Elephant and Rhino Specialist Group Technical Bulletin No 1:1–16. Gartlan S. 1989. La conservation des écosystèmes forestiers du Cameroun. IUCN,Gland, Switzerland. 186 p. Hall J, Nkwi P. 1993. Proposal for the south-eastern Cameroon component of the Global Environment Facility Project for Cameroon. Lake Lobeke strategic planning mission summary report. (submitted by Jefferson S. Hall et Bryan K. Curran, 8 March 1993 WCI). Harrison M, Agland P. 1987. A draft proposal for the designation of three new national forest parks. Dja River Films Ltd. Letouzey R. 1985 Notice de la carte phytogéographique du Cameroun au 1/500000è. Institut de la Carte International de la Végétation. UNEP, Toulouse, France. [MINEF] Ministère de l’Environnemnt et Forêts. 2001. Le Cameroun au sommet : sommet mondial pour le développement durable. MINEF, Yaoundé. 15 p. Nzouango D. 1994. Boumba-Bek. Etudes préliminaires. World Conservation Society. Yaoundé, Cameroon. 28 p. Simon NS, Adams RJ, Jenkins MD. 1989. Biodiversity in sub-Saharan Africa and its islands: conservation, management and sustainable use. A contribution for the Biodiversity Conservation Strategy Programme. IUCN, Gland. 242 p. Stromayer KA, Ekobo A. 1991. Biological surveys of southeastern Cameroon. Wildlife Conservation International, Bronx, New York. Usongo L, Noupa P. 2001. The way forward: Boumba-Bek and Nki proposed parks, Southeast Cameroon. Vision paper. WWF Jengi SE Forest Project, Youndé, Cameroun. 6 p. 81 Choudhury Threats to the greater one-horned rhino and its habitat, Pabitora Wildlife Sanctuary, Assam, India Anwaruddin Choudhury The Rhino Foundation for Nature in NE India, c/o Assam Co. Ltd., Bamunimaidam, Guwahati 781 021, Assam, India; email: [email protected] Abstract Pabitora Wildlife Sanctuary in Assam, north-eastern India, has the highest density of the Indian one-horned or greater one-horned rhinoceros, Rhinoceros unicornis, anywhere in its range. This area, dominated by moist savanna grasslands, was designated a wildlife sanctuary in 1987. With around 80 rhinos, Pabitora is an important habitat for this endangered species. However, the area is facing serious threats such as encroachment, road construction, overgrazing, poaching, high floods and increasingly heavy tourism. This paper discusses these threats and issues and suggests solutions. Additional key words: Rhinoceros unicornis, grassland, flooding, poaching, siltation Résumé Le Sanctuaire de la Faune de Pabitora en Assam, au nord-est de l’Inde, comprend la plus forte densité de rhinocéros unicornes indiens (Rhinoceros unicornis) de toute son aire de répartition. Cette région, dominée par une savane de prairies humides, a été désignée comme sanctuaire pour la faune sauvage en 1987. Avec quelque 80 rhinos, Pabitora est un habitat important pour cette espèce en danger. Pourtant, la région fait face à de graves menaces comme l’empiètement, la construction de routes, le surpâturage, le braconnage, les inondations et un tourisme de plus en plus envahissant. Cet article discute ces menaces et problèmes et suggère des solutions. Mots clé supplémentaires : Rhinoceros unicornis, prairies, inondations, braconnage, sédimentation Introduction Pabitora Wildlife Sanctuary (26°14N–16N N, 91°57N– 92°05N E) in Assam, north-eastern India, is a known stronghold of the endangered Indian one-horned or greater one-horned rhinoceros, Rhinoceros unicornis (Vigne and Martin 1984, 1998; Choudhury 1985, 1991, 1997; Talukdar 2000; ). It also has the highest density of R. unicornis anywhere in its range. Located in Morigaon District (with a small portion also in Kamrup Metropolitan District) on the south of the Brahmaputra River, it covers 38.8 km2 (fig. 1). The habitat is dominated by moist savanna grassland with patches of woodland and marshy pockets, and a large hillock. Useful information on Pabitora, including the now-shelved project of introducing the critically en- 82 dangered subspecies of the brow-antlered deer, Cervus eldi eldi, is found in Choudhury (1987, 1989a,b,c,d, 2002), Rahmani et al. (1988) and Barua (1994). Pabitora has an interesting history; in the early part of the last century it was neither a reserve forest nor an identified habitat of rhinoceros. The area was used for grazing domestic cattle and buffaloes and it was recorded in the revenue department as ‘professional’ and ‘village’ grazing reserves. In the 1960s the villagers of Raja Mayong, Lunmati and Burha Mayong demanded that the area be declared a reserve forest to prevent migrants from encroaching in it (Bengali Moslems and Hindus from the former eastern Bengal and East Pakistan, now Bangladesh) and also to protect the rhinos. In 1961/62 the Nagaon For- Pachyderm No. 38 January–June 2005 Threats to the greater one-horned rhino, Pabitora, India A. Choudhury est Division confirmed the presence of a few rhinos in the area. The villagers extended their full support and continued to urge the government to protect the area. In 1971 it was declared a reserve forest; in 1987 it was designated a wildlife sanctuary. The de facto area of the sanctuary was about 16 km2 of grassland interspersed with wetland and a large patch of woodland. At the time of final notification in 1998, the sanctuary area was increased to 38.8 km2, which included 15.85 km2 of de facto sanctuary, 12 km2 of Raja Mayong Reserve Forest (RF) and about 11 km2 of other government land, called khas land by the revenue department (fig. 1). Inclusion of the additional area was finalized after all the villagers’ claims and rights over the land had been settled. Rhinos in Pabitora do not share their habitat with many other large mammal species. Those present include the wild pig (Sus scrofa), jackal (Canis aureus), and feral water buffaloes, among which is probably at least one pure wild bull of Bubalus arnee (= bubalis) that came during high floods. The sanctuary is known for its rich birdlife. In the Raja Mayong hills, leopard (Panthera pardus), muntjac (Muntiacus muntjak) and rhesus monkey (Macaca mulatta) are found besides various other smaller species. This article discusses serious problems faced by this important rhino habitat and presents possible solutions. Methods Since the 1980s I have been visiting this wilderness, which is only about 48 km by road from Guwahati. I obtained data presented here at first hand as a researcher, as an activist of The Rhino Foundation for Nature in North-East India, and as an official of the Department of Environment and Forests of the Assam government. Census figures are departmental counts carried out by direct counting from elephant back, in which I was also involved either directly or indirectly. The problems The main objective of this work was to highlight the serious problems that have threatened a globally important rhino habitat and their possible solutions. Area of the sanctuary: Raja Mayong RF is actually a rocky hillock unsuitable for rhinos although a few animals occasionally climb the slopes, sometimes ending up dead among the rocks (see photo). Khas land on the other hand is excellent for rhinos as it is in the floodplain and consists of wetland and grassy tracts. By the time formalities demarcating the sanctuary were finalized, much of the khas land was already under human occupation, both permanent and temporary, by farmers in adjacent villages, who had started wet paddy cultivation and intensive fishing. As a result this area of 11 km2 cannot be used as a sanctuary although it is designated as such. Human occupation and intensive cultivation as well as the jutting shape near Mayong of the Murkata part of the sanctuary make it difficult to reclaim this area. But the Kamarpur khas land, the area that connects the original sanctuary with Raja Mayong RF, should be cleared of human activity (fig. 1). Growing rhino population: Table 1 shows the increase in rhino population in Pabitora. The rhino This rhino got caught in the rocky terrain—and, unable to extricate itself, died. Pachyderm No. 38 January–June 2005 83 Choudhury habitat remains the original 15.85 km2 for around 80 rhinos. The increase in area has not helped the animals. The rhinos regularly stray out at night to nearby fields, which include both khas land that is part of the sanctuary and private land. While the growing number of rhinos is certainly not a problem, as additional rhinos may be translocated, the lack of additional habitat and the failure to get hold of the added area of the sanctuary are major issues. Roads: An all-weather road (closed only during high floods) passes through the western boundary of the de facto sanctuary. An old, infrequently used road being reconstructed will pass through the sanctuary at three places, posing a threat to the sanctuary and its rhinos. This road will not only disturb the animals but may also cause a few accidents with them. It will, however, be an important road, connecting Guwahati with Morigaon and Nagaon through Chandrapur and Mayong, and it will be shorter than the existing road. The bridge over Kolong River is complete and once this road is constructed visitors to Pabitora from Guwahati will use it. Busy traffic is expected on this road. Mortality: Even the slightest negligence can result in rhinos getting killed. For instance, a mother 2 Hiloikhunda 3 Hatimuria Sanaka Kajoli Haduk Tamuliduba Jukdol Garanga Pagladuba Duboritoli 1 Brahmaputa Brahmaputa River river 4 5 6 6 Murkata MAYONG RAJA MAYONG HILLS Burha Mayong Sildubi 1 To Guwahati Tuplung 2 3 Kamarpur To Morigaon Haduk Kholabhuyan Jukdol Kamarpur 4 5 Pabitora Kukuwari To Sonapur Thengbhanga Kukuwari INDIA N 0 2 km Approx. Assam sanctuary boundary Diprang Nekera main sanctuary MANAS KAZIRANGA PABITORA added areas river road forest range office anti-poaching camp Main habitat types grassland woodland waterbody (including marsh and swamp) Figure 1. Pabitora Wildlife Sanctuary showing habitat types. 84 Pachyderm No. 38 January–June 2005 Threats to the greater one-horned rhino, Pabitora, India Table 1. Estimated number of rhinos in Pabitora Wildlife Sanctuary according to census Year 1961–62 1969–70 1993 1995 1999 2004 Adult Subadult Male Female Unidentified Male – – 18 11 17 14 – – 21 28 26 33 – – 1 3 0 0 – – 1 3 7 5 Calf Total Female Unidentified – – 2 2 5 4 – – 2 13 0 2 – – 11 9 19 21 a few 20 56 69 74 79 Source of 1961–62 and 1969–70 counts: P.C. Gogoi, Working Plan for Nagaon Division – no detail is available Table 2. Mortality of rhinoceros in and just outside Pabitora Wildlife Sanctuary Year Poaching 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 (to 25 Oct.) 2 3 4 2 1 3 4 4 2 5 3 4 6 2 0 3 2 0 Natural death 3 5 3 2 1 2 1 2 1 2 2 2 0 1 0 2 3 5 Total 5 8 7 4 2 5 5 6 3 7 5 6 6 3 0 5 5 5 Source: Department of Environment and Forest and The Rhino Foundation for Nature in NE India A. Choudhury rhino with a calf were electrocuted in 2003 during a few hours of slackness by staff. The authorities immediately punished the staff and later suspended the ranger for his overall negligence. Although rhino poaching has lessened in recent years, it still remains a major problem (see photo). Poachers were also nabbed carrying a horn. Table 2 shows the number of rhinos that have died through poaching and other causes (usually recorded as ‘natural’) between 1987 and 2004. Flooding: Annual flooding is essential for the survival of the alluvial grasslands of Pabitora, but periodic high floods are detrimental. In the 1990s, the worst flood was recorded in 1998; so far since the turn of the century, it was in July–August 2004. During such high floods, the entire de facto sanctuary reels from the effects of the floodwaters. Many rhino calves perish and the grassland is damaged. Two rhinos drowned in 1998 and four in 2004. And when animals move out of the sanctuary during this period they provide poachers with the opportunity to strike. Rhino poaching, although declining, still takes its toll. Pachyderm No. 38 January–June 2005 85 Choudhury Siltation: Although siltation is a natural phenomenon, due to the small size of the sanctuary a major crisis could result during the drier winter months if waterbodies filled up. In the early 1990s Tamuliduba was a fine wetland, but it has now become a seasonal marsh that virtually dries up in late winter. For rhinos, the year-round presence of waterbodies is essential. Vanishing grassland: Pabitora’s savanna grassland is vanishing fast. Grazing by domestic cattle and buffalo is the main pressure on the grassland; next is villagers’ illegal collection of grass and reeds for building material and for sale. Since the 1998 high floods, the grassland has not recovered as prolonged waterlogging caused much damage. Subsequent heavy grazing by domestic stock, numbers of which are also increasing yearly and now probably are 4000–5000 head, also has not allowed it to regenerate. The situation has become so precarious that during the pre-monsoon period rhinos take shelter in the woodland in the centre of the sanctuary. The decay in grassland habitat has also resulted in a sharp decline in swamp francolin (Francolinus gularis) population (Choudhury 2000). Fast-growing tourism: Due to Pabitora’s closeness to the city of Guwahati, the capital of Assam State with a population of almost 1 million people, a large number of domestic day tourists throng the sanctuary every winter. Most of these visitors are picnickers who like to play loud music, causing noise pollution, and who leave behind a heap of refuse including plastic materials. Up to a hundred buses were counted on Sundays during winter. Illegal fishing: While illegal fishing is a perennial problem in Pabitora, a large wetland, Garanga, is used by a commercial contractor who got the lease from the Fisheries Department against the provisions of the Wildlife (Protection) Act 1972 as amended in 2002. Suggested solutions Area of the sanctuary: Due to stringent provisions of the Wildlife (Protection) Act, modifying the boundary of the sanctuary will not be an easy task. Hence, the authorities must take urgent steps to take full control of the 11 km2 of khas land that was designated as part of the wildlife sanctuary. The part of khas land that is intensively cultivated and inhabited should be treated as an intensive tourism zone in the management plan and some of the villagers then made stakeholders in tourism projects so that they do not lose what they have invested but stand to profit from such projects. Privately 86 owned elephants may also be allowed to carry tourists in the tourism zone. The small encroachment in the Diprang area should be sorted out by vigorously pursuing the case that is now in the court. Growing rhino population: A detailed ecological study is needed to determine the carrying capacity of Pabitora. The 16-km2 area of de facto sanctuary cannot support the growing number of animals for an indefinite period. Hence a comprehensive plan should be mapped for translocating rhinos, such as to Laokhowa and Burhachapori Wildlife Sanctuaries and Manas National Park. The successful translocations in Nepal may be taken as examples. Roads: The all-weather road that passes through the western boundary of the original sanctuary should be realigned along the existing road through the southern and eastern boundary for vehicle traffic. The road from Guwahati that will pass through the sanctuary at three places should be redirected at the north-east corner of the sanctuary so that it does not cut off the small area on the other side of the road near Kholabhuyan. For a small sanctuary like Pabitora, even a tiny chunk is important. Little probably needs to be done about where the road cuts across at Burha Mayong and Murkata as the impact on the sanctuary will not be much. Speed-breakers and checkgates should be placed at suitable places on the road. Law enforcement: Exemplary punishment of erring staff should be made as the need arises, as was done in 2003 after the electrocution of two rhinos so that there is no room for complacency. The anti-poaching network should be strengthened and provided with better arms; personnel should be trained like the armed forces in how to use arms. Non-governmental organizations such as The Rhino Foundation for Nature in NE India and Aaranyak should continue to maintain their network of informers to supplement government efforts on anti-poaching. Flooding: The high floods of 1998 and 2004 showed the important role a high platform of artificially raised ground plays in saving marooned animals. However, the present mounds were not properly constructed and are not sufficiently high. There should be strict monitoring of such vital works and new highlands should be constructed at least a metre higher than the highest flood level. Responsibility should be predetermined, in case of any negligence on such issues. No measures need to be taken to control floods as the alluvial grasslands depend on them. Siltation: Rhinos need a permanent waterbody as Pachyderm No. 38 January–June 2005 Threats to the greater one-horned rhino, Pabitora, India well as marsh in winter when the seasonal wetlands become dry. To save the perennial wetlands from drying out in winter, selected sites should be desilted yearly. As Pabitora is a small area, such regular habitat manipulation is possible. Vanishing grasslands: For the savanna grassland to recover to its pre-1998 condition and for its longterm survival, electric fencing may be the only option. The Rhino Foundation for Nature in NE India prepared a proposal a few years back, which also had the endorsement of the Forest Department, but it was not funded, mainly because it proposed only partial fencing. But the reality is that fencing the entire area is neither possible nor necessary. Cattle from a particular village enter from the boundary contiguous with that village or through neighbouring areas only. And the pressure is not equal from all sides. If the main entry sides are fenced off, other areas such as water areas where fencing is not feasible could be guarded. The purpose of fencing should not be to completely stop movement but to halt degradation of the grassland through overgrazing. Fast-growing tourism: Tourism should not be discouraged even where it causes some damage; areas with negligible or no tourism become more vulnerable to poaching, encroachment, illegal felling and illegal fishing, and staff become inactive and complacent. Villagers living at the fringe of the sanctuary do not realize the importance of the area nor do they presently benefit in any way. Tourism acts as a monitoring mechanism; the idea is to regulate it. Areas such as Murkata should be developed as intensive tourism zones where village families whose farms would be affected can engage in economic activities such as providing tourists with elephant rides and decent accommodation, in homes or tented camps. Stakeholders should include the private sector, as wholly government-controlled tourism projects may not include or encourage fringe villagers. Sites for picnickers should be identified on the edges of the tourism zone. Illegal fishing: To curb illegal fishing in the sanctuary, surprise visits by senior officials are recommended. The problem of the illegal lease for commercial fishing in Garanga should be sorted out permanently in a court of law. Discussion Pabitora is the second most important rhino area in India, after Kaziranga, and its problems must be Pachyderm No. 38 January–June 2005 viewed seriously and addressed before it is too late. Areas such as Murkata and Kamarpur khas land were scarcely occupied in 1987 when Pabitora was made a reserve, but delays in disposing of claims and rights for a decade have resulted in encroachment and intensive human activity. Converting additional areas, especially the khas land portion, into tourism zones with villagers as stakeholders appear to be the only likely option. The time has come for a comprehensive translocation plan to move a specified number of rhinos every year. The roads should be realigned for the greater long-term interest. Anti-poaching networks should be strengthened and NGO support should continue. Since the mid-1990s, various NGOs have supplemented government efforts, which have improved anti-poaching activities. Plans to tackle high floods by constructing areas of high ground should be on management’s agenda. For the long-term survival of the grasslands, select stretches should be fenced. In 1993–94, due to stringent measures by sanctuary authorities, the number of cattle grazing in the sanctuary was brought down from 4000 to 300 a day (Barua 1994). Tourism should be encouraged in a big way, and it should involve the private sector. Illegal fishing is often a bone of contention between villagers and the authorities, as the former feel they have been forced to give up fishing rights, only to see others fish illegally. Hence there is need for stringent monitoring. Pabitora is surrounded by at least 21 villages with more than 10,000 inhabitants, so the threat of encroachment is constant. The Department of Environment and Forests recently prepared a five-year management plan (Bora 2003); although modifications may be required from time to time, its implementation should be vigorously pursued. Acknowledgements I wish to thank the following persons: the late Nagen Sharma (Forest minister, 1996–2000), Pradyut Bordoloi (Environment and Forest minister since 2001), Anne Wright for continued encouragement, L. Rynjah and P.P. Varma (both principal secretaries of the Forest Department), S. Doley and M.C. Malakar (both chief wildlife wardens), S. Islam, C.K. Bora and Sanjeeb Bora (all district forest officers), P. Barua, Mrigen Barua, P. Deka and R. Das (all range officers), Capt. T. Phukan of the Bhoreli Angler’s Association, Moloy Barua of Early Birds, Aneisha Sharma, Ratul 87 Choudhury Talukdar, Hakim, Montu Nath and the late Sakul Boro. Many of the inputs given by the Rhino Foundation were supported by the US Fish and Wildlife Service, Fauna and Flora International, and the Rhino Rescue Trust, and all deserve mention. References Barua AK. 1994. The survey of the fringe villages of Pabitora Wildlife Sanctuary and its eco-development management plan. Guwahati: Department of Forests, Government of Assam. Bora CK (comp). 2003. Management plan of Pobitora Wildlife Sanctuary (2003–04 to 2007–08). Nagaon: Nagaon Wildlife Division, Government of Assam. Choudhury AU. 1985. Distribution of Indian one-horned rhinoceros. Tigerpaper 12(2):25–30. Choudhury AU. 1987. A day in Pobitora. The Sentinel (Guwahati): 26 April 1987. Choudhury AU. 1989a. Pobitara, Assam’s Rhino Reserve. The India Magazine 9 (August):46–54. Choudhury AU. 1989b. Deer challenges rhino in Pobitora. The Sentinel (Guwahati): 12 November 1989. 88 Choudhury AU. 1989c. Rhino vs. deer in Pobitora. WWF– India Quarterly 11(1):11. Choudhury AU. 1989d. Conflict at Pabitora. Oryx 24:68. Choudhury AU. 1991. Indian rhino, what’s that? Telegraph Colour Magazine 27(October):6–10. Choudhury AU. 1997. Checklist of the mammals of Assam. rev. 2nd ed. Gibbon Books & Assam Science, Technical and Environment Council, Guwahati. Choudhury AU. 2000. The birds of Assam. Gibbon Books and WWF-India NERO, Guwahati. Choudhury AU. 2002. Saved but not ‘safe’: the story of the Indian rhino. Environ 8(3):39–47. Rahmani AR, Narayan G, Rosalind L, Sankaran R, GanguliLachungpa U. 1988. Bengal florican Houbaropsis bengalensis: status and ecology. Annual Report no. 3. Bombay Natural History Society, Bombay. Talukdar BK. 2000. The current state of rhino in Assam and threats in the 21st century. Pachyderm 29:39–47. Vigne L, Martin EB. 1984. The greater one-horned rhino of Assam threatened by poachers. Pachyderm 18:28–43. Vigne L, Martin EB. 1998. Dedicated field staff continue to combat rhino poaching in Assam. Pachyderm 26:25–39. Pachyderm No. 38 January–June 2005 The royal hunt in the Nepalese terai in 1911 HISTORY The royal hunt of tiger and rhinoceros in the Nepalese terai in 1911 Kees Rookmaaker,1* Barbara Nelson2 and Darrell Dorrington2 1 Rhino Resource Center, c/o IUCN Species Survival Programme, 219c Huntingdon Road, Cambridge CB3 0DL, UK; email: [email protected]; *corresponding author 2 Asia Pacific Collections, Menzies #2, Australian National University; Canberra ACT 0200, Australia email: [email protected] Abstract King George V took part in a large organized hunt in the Chitwan area of Nepal from 18 to 28 December 1911. During these days, the king and his party killed 18 rhinos, 39 tigers and 4 sloth bears. A young live rhino was presented to them by the prime minister of Nepal and transported to the London Zoo. The events were captured in photographs found in three albums preserved in the library of the Royal Commonwealth Society (University of Cambridge) and the Australian National University. Résumé Le roi George V prit part à une grande chasse organisée dans la région népalaise de Chitwan du 18 au 28 décembre 1911. Pendant ces 11 jours, le roi et sa suite ont tué 18 rhinos, 39 tigres et 4 ours lippus. Le Premier ministre du Népal leur présenta aussi un jeune rhino vivant qui fut transporté au Zoo de Londres. Ces événements furent immortalisés par des photographies qui sont rassemblées dans trois albums qui sont conservés à la bibliothèque de la Royal Commonwealth Society (Université de Cambridge) et à l’Université Nationale Australienne. Introduction An earlier survey of historical sources about rhinos in Nepal by Rookmaaker (2004) recorded shoots organized for British royalty in 1906 and 1921, but rather glossed over a similar event in 1911. When details about a photo album with pictures of this shikar became available, a search was made for additional written sources. These were found to include narratives of the royal visit to India and Nepal in Fortescue (1912), anonymous (1914) and Day (1935, largely copied from Fortescue). This enabled us to provide a daily account of the 1911 shoot in Nepal, which focuses on information about rhinos and tigers found in the Nepalese terai. It must be said that Pachyderm No. 38 January–June 2005 this summary presents rather depressing reading, being a record of who shot and killed what and when. Public opposition to such events was probably largely confined to royal hunting in England (Taylor 2004). We present the facts as found in contemporary sources, give the total of the animals killed, and introduce two sets of photographs portraying this royal excursion to southern Nepal. The imperial visit to Nepal in 1911 George V (1865–1936), king of Great Britain and Ireland from 1910, went to India in 1911 to be crowned Emperor of India in a great durbar in Delhi on 12 December 1911. After the ceremony, a visit to Nepal 89 Rookmaaker et al. to participate in a hunt of tiger and rhinoceros was organized for his pleasure. Although the king of Nepal, Maharajadhiraja Prithivi Bir Bikram Shah Deva (1875–1911), suddenly died on 11 December 1911, it was decided to proceed with the preparations. Hence on 16 December 1911, King George and Queen Mary took different routes from Selimgarh station on the outskirts of Delhi. The king travelled by train and boat via Patna and Bankipore to Bikna Thori, where a temporary station was erected on the Bengal and North-Western Railway (see fig. 1 for localities mentioned in the text). Arriving on Monday, 18 December, at 10 a.m., the king was met by the prime minister and actual ruler of Nepal, HH Projwala-Nepala Taradisha Sri Tin Maharaja Sir Chandra Shumsher Jung Bahadur Rana (1863–1929), together with his two sons and British Resident in Nepal Lieutenant-Colonel John Manners-Smith (1864–1920). Here the hunt started and we summarize the daily events from the three books mentioned in the introduction. Monday, 18 December 1911 From the station of Bikna Thori, the king proceeded by car through the valley of the Rui River on a 13mile road constructed for the occasion. Here he was met by General Maharajkumar Sir Baber Shumsher Jung Bahadur Rana (1888–1960), the second son of the prime minister. The king and his party went after elephants in the jungle for about 1 mile (1.6 km). At this time the king shot two tigers, which had been secured within a ring formed by the howdah elephants. After motoring for another 19 miles and taking a short break for lunch, the royal party arrived at a second ring of elephants. Lord Durham shot a tiger, the king dispatched a pair of rhinos, and Lord Durham and Lord Annaly between them shot a third rhino. At 5.30 p.m. they reached the camp at Sukhibar (Sakhi Bar) on the south bank of the River Rapti. The setting was wonderful: ‘with a glorious climate, wonderful scenery, and always to the north the incredible panorama of the eternal snows towering into the sky’ (Smythies 1961). In the camp a bungalow fitted with electric light had been built for the king, surrounded by the tents of his party. To give an idea of the operation: ‘outside the fencing of the camp were various smaller encampments for the Residency Escort, the motor-cars, stables, taxidermists, hospital, laundry, the post and telegraph offices, and other subsidiary services. The camp of the maharaja, who had with him some members of his family as well as his staff and senior officers, was situated also on the river bank a little lower down and behind it, hidden in the jungle, Maruwa Ghat T E R A I Belani Narayangarh Chilha Tikoli Meghauli Nandpur Tamaspur Bharatpur Narayani River N E P A L Bharatpur R O Y A L C Patlahara Khargauli Debichaur Jhawani r Rive KASRA SUKHIBAR pti a R H I T W A N N A T I O N Tribeni Ghat Bhaisalotan Bhagaufa A L P A R K Harta Rui Rive r I N D I A Bikna Thori Bikna Thori Hamatanr Gobardhana Figure 1. The terai region in Nepal, showing places mentioned in the text. 90 Pachyderm No. 38 January–June 2005 The royal hunt in the Nepalese terai in 1911 was the large encampment of His Excellency’s followers, who numbered twelve thousand, besides six hundred elephants with two thousand attendants’ (anonymous, 1914: 231). Durham, Lord Charles Fitzmaurice, Sir Derek Keppel, Sir Colin Keppel and Sir Henry McMahon among them killed seven tigers and a Himalayan bear. Saturday, 23 December 1911 Tuesday, 19 December 1911 Every night baits were put out for tigers over a large area of the country. News of a tiger on a kill reached the camp just after mid-day. A party went out and the king shot the animal. In the afternoon, Sir Charles Cust shot another tiger, while Sir Colin Keppel and Sir Horace Smith-Dorrien each killed a rhino. Wednesday, 20 December 1911 On this day three separate parties went out, two looking for tiger and one for rhinoceros. In the first ring, the king soon shot a tigress. In the second ring, the hunters disturbed a rhino cow and calf, which charged straight at the elephants and broke out of the ring without being hurt. After lunch it was found that the next ring contained no fewer than four tigers ‘roaring and snarling in a blood-curdling chorus, the tigers charged madly from side to side while the surrounding elephants trampled and trumpeted and the mahouts screamed and shouted’ (Day 1935: 99). The king killed all four tigers. On the way back to camp, a big bull rhino came out of the bush and the king shot it with two shots. The other parties were also successful: Captain Godfrey Faussett and Sir Colin Keppel each shot one tiger, Captain Godfrey Faussett and Sir Horace SmithDorrien each shot a bear, and the Duke of Teck shot a rhino. Elsewhere, a rhino pursued the elephant on which Major Wigram was riding, but abandoned the chase after half a mile (0.8 km) and ran away. This day was taken up by a move to a second shooting camp located at Kasra, 8 miles (12.8 km) further up the River Rapti. It is likely that the size of the first camp had disturbed the wildlife in the area so much that most animals had fled to other areas of the terai. Sunday, 24 December 1911 In the morning, there was a divine service led by Rev. J. Godber, domestic chaplain to the bishop of Calcutta. In the afternoon, the king went with General Kaisar Shumsher to inspect a collection of animals from Nepal the maharaja had presented to him. There were over 70 kinds of animals, including a young elephant, a rhino calf, and a wild ass; these animals were sent to the zoo in London. Various items of Nepalese art were also presented, and they are now in the Victoria and Albert Museum, London. In the evening, the king invested the Maharaja Sir Chandra Shumsher Jung with the insignia of a Knight Grand Commander of the Royal Victorian Order. Monday, 25 December 1911 In the afternoon, after travelling 3 miles from camp (4.8 km) the king killed a tiger measuring 9 feet 6 inches (290 cm). He then went after rhinos and first shot a female, then a second animal after lunch. A party led by the Duke of Teck shot three tigers and a rhino. Tuesday, 26 December 1911 Thursday, 21 December 1911 In the morning, the king killed four tigers and a bear. Captain Faussett was pursued by a bull rhino, which he later killed. The game in the area was already disturbed, obvious because only one of the 60 baits put out the previous evening had been touched by a tiger. Lord Durham shot one tiger and after a long beat the king finally managed to shoot a rhino. Friday, 22 December 1911 Wednesday, 27 December 1911 On this day, the king shot three tigers. When a big bull rhino suddenly ran out of the grass, shots by Sir Charles Cust and the king missed it, but he was killed by Captain Faussett. On separate occasions Lord Pachyderm No. 38 January–June 2005 The king shot one tigress, said to be his 20th since his arrival in Nepal. 91 Rookmaaker et al. Thursday, 28 December 1911 valley of southern Nepal was 18 rhinos (Rhinoceros unicornis), 39 tigers (Panthera tigris) and 4 sloth bears (Melursus ursinus). Of these, King George V shot 8 rhinos, 21 tigers and 1 bear. In the narratives summarized above, only 13 rhinos and 31 tigers are accounted for, but we assume that this is an oversight and that the king killed none of the other animals. The persons credited with the various animals are listed in table 1, in alphabetical order after the king. The dates in December when the animals were shot are given followed by the number of that species on that day. On this last day of the hunt, the king and the Duke of Teck simultaneously shot one tiger. After lunch, farewells were said and the party travelled 12 miles (19 km) by car to the train, which was waiting for them at Bikna Thori. On the way, the king shot his last (21st) tiger. Animals obtained in 1911 According to Fortescue (1912) the total number of animals shot during this 10-day hunt in the Chitwan Table 1. The Great Hunt: 18–28 December 1911, Chitwan area, Nepal The hunters and their titles Rhinos shot and dates Tigers shot and dates Bears shot and dates King George V (1865–1936) 6: 2 on 18th; 1 on 20th; 2 on 25th; 1 on 26th 16: 1 on 19th; 5 on 20th; 4 on 21st; 3 on 22nd; 1 on 25th; 1 on 27th; 1 on 28th 1 on 21st Lord Annaly: Sir Luke White, 3rd Baron Annaly of Annaly and Rathcline (1857–1922), permanent Lord-in-Waiting to George V from 1910 to 1921 1 on18th; shot with Lord Durham — — Sir Charles Leopold Cust (1864–1939) — 1 on 19th — Lord Durham: John George Lambton, 3rd Earl of Durham (1855–1928) — 4: 3 on 22nd; 1 on 26th 1 on 22nd Captain Faussett: Captain Sir Bryan GodfreyFaussett (1863–1945), captain in the Royal Navy, Equerry to George V 2:1 on 21st; 1 on 22nd 1 on 20th 1 on 20th Lord Charles Fitzmaurice: Sir Henry Charles Keith Petty-Fitzmaurice (1845–1927), 5th Marquess of Lansdowne — 1 on 22nd — Admiral Sir Colin Keppel (1862–1947): admiral in the British navy (retired 1932). Commander of the HMS Medina on the voyage to India with HM King George V for the King Emperor’s durbar in 1911 1 on 19th 2: on 20th; 1 on 22nd — Sir Derek William George Keppel (1863–1944): Deputy Master of the Household to HM King George V between 1910 and 1912 — 1 on 22nd — Sir Henry McMahon (1862–1949): British High Commissioner in Cairo — 1 on 22nd — General Sir Horace Lockwood Smith-Dorrien (1858–1930) 1 on 19th — 1 on 20th 4: 3 on 25h; 1 on 28th — Duke of Teck (1868–1927): Adolphus Charles 2: 1 on 20th; 1 Alexander Albert Edward George Philip Louis on 26th Ladislaus, 2nd Duke and Prince of Teck (from 1900). In 1917 adopted the surname Cambridge. Personal Aide-de-Camp to the Sovereign, 1910–1914 92 Pachyderm No. 38 January–June 2005 The royal hunt in the Nepalese terai in 1911 How the various trophies were distributed is unknown. The rhino presented to the king on 24 December 1911 was, with the rest of the collection, for some time kept in the Alipore Zoological Gardens in Calcutta (Basu 1912). It left Calcutta on 1 April 1912 on board the SS Afghanistan and lived in the Zoological Gardens in London from 21 May 1912 to 2 November 1921 (Edwards 1996: 130; Rookmaaker 1998). Australian National University, Canberra, Australia (http://images.anu.edu.au/tigers), associated with King George’s hunt in Nepal in 1911. This led to the discovery of two more albums of photographs in the library of the Royal Commonwealth Society (RCS). Details of these three albums are presented here. Finally, there is a coloured engraving showing George V seated on an elephant hunting a rhino, published in the 31 December 1911 issue of Le Petit Journal (no. 1102) issued in Paris, France (fig. 2). Illustrations of the hunt In 1950, Queen Mary presented a large and valuable collection of books, scrapbooks and 29 photoWhen Rookmaaker (2004) submitted his paper, he graphic albums relating to British India and the royal found it almost impossible to find any historical il- family to the RCS in London, which the library of lustrations of the rhinoceros in Nepal. He then saw a the University of Cambridge obtained in 1992. number of relevant photographs on the website of the Handlists of the contents of the collection are available in the library. One of the photograph albums (QM 21) is a large, oblong volume (52 x 37 cm), beautifully leather bound, entitled ‘H.I.M. The King–Emperor of India’s Shooting in the Nepalese Terai, December, 1911’. It has 41 leaves with 278 black-and-white photographs, which are numbered but not otherwise annotated. This appears to be a chronological record of all events from the arrival of King George V in Nepal until his departure 10 days later. It appears that every tiger and rhino seen or shot in this period is figured, as well as the collection of live animals given to the king. There are 10 photos of dead rhinos (nos. 45, 92, 125–128, 206, 235, 242, 243), one of a rhino hidden in the grass (no. 219), one of the captured young rhino given to the king (no. 165), and one of a collection of trophies including a mounted baby rhino (no. 177). This album QM 21 represents the most comprehensive record of the king’s shoot in Nepal in 1911 that has come to light. The second photographic album in the RCS (QM 20) is a large, oblong volume (42 x 31 cm) entitled ‘Indian Tour 1911–1912’. It has photographs taken during the whole tour, with those taken during the hunt in Nepal numbered 196 to 333; none are annotated. Figure 2. Engraving (in colour) of the king’s hunt in Nepal, from Le Petit Journal, 31 December 1911. There are eight photographs of dead Pachyderm No. 38 January–June 2005 93 Rookmaaker et al. rhinos (nos. 244, 261, 294, 296, 299, 301, 304, 330) and one of the captured young rhino (no. 165), mostly different from those in QM21. For a number of reasons, these photographs cannot be reproduced at present. The photographic album in the Australian National University consists of 16 pages and 50 photographs (signature: MENZIES v.lge rare bk + 2108458; anonymous 1912). It appears incomplete with the back cover missing. The front cover bears the company name Herzog & Higgins, Mhow (Central India) and the title ‘His Imperial Majesty’s shoot in Nepalese Terai, December 1911’. Inside the front cover it is stated ‘Bound at the Caxton Works, Bombay’. Herzog & Higgins operated a photographic studio from the 1890s to the 1920s at Mhow, a British military cantonment town 22 km from Indore, Madhya Pradesh, India. The album was donated by Dr U.N. Bhati. He had discovered it in Madhya Pradesh, in the rural home of distant relatives, Mr and Mrs Parbinder Singh, who had worked for the former maharaja and maharani of Ratlamand, and who gave him the album. Although it is unlikely that this album is unique, no identical copies have been found in the course of this research. There are five photographs with a rhino, all of which are also found in the RCS album QM 21. One shows the animal running in the grass; three were taken after the rhino was shot (figs. 3–5), while the last one shows the young animal that was presented to the king and that was later shipped to London (fig. 6). There are several photos with tigers (fig. 7) and other scenes of the people involved. All photographs in the album are available on the Internet on the website of the Australian National University (http://images.anu.edu.au/tigers). Figure 3. Shikari flays a dead rhinoceros (His Imperial Majesty’s Shoot in Nepalese Terai, page 14, photograph 4; also RCS QM 21, no. 243). 94 Pachyderm No. 38 January–June 2005 The royal hunt in the Nepalese terai in 1911 Figure 4. Hunting party inspects the head of a slain rhinoceros (His Imperial Majesty’s Shoot in Nepalese Terai, page 14, photograph 3; also RCS QM 21, no. 128). Figure 5. Hunters with slain rhino (His Imperial Majesty’s Shoot in Nepalese Terai, page 5, photograph 4; also RCS QM 21, no. 235). Pachyderm No. 38 January–June 2005 95 Rookmaaker et al. Figure 6. Rhino presented to King George V by Maharaja Sir Chandra Shumsher Jung Bahadur Rana in December 1911, later transported to the London Zoo (His Imperial Majesty’s Shoot in Nepalese Terai, page 7, photograph 2; also RCS QM 21, no. 165). Figure 7. Four tigers and a deer shot during the hunt (His Imperial Majesty’s Shoot in Nepalese Terai, page 14, photograph 2). 96 Pachyderm No. 38 January–June 2005 The royal hunt in the Nepalese terai in 1911 Acknowledgements The staff responsible for the collection of the Royal Commonwealth Society in the library of the University of Cambridge have helped locate the photographs relating to the Nepalese hunt. Comments by reviewers have helped improve the contents of this paper. The work of the Rhino Resource Center is supported by the International Rhino Foundation and SOS Rhino. References Anonymous. 1912. His Imperial Majesty’s shoot in Nepalese Terai, December 1911. Herzog & Higgins, Mhow, India. Anonymous. 1914. The historical record of the imperial visit to India, 1911, compiled from the official records under the orders of the Viceroy and Governor-General of India. John Murray, for the Government of India, London. Basu B. 1912. Report of the Honorary Committee for the Management of the Zoological Garden, for the year 1911–12. Bengal Secretariat Depot, Calcutta. Pachyderm No. 38 January–June 2005 Day JW. 1935. King George V as a sportsman: an informal study of the first country gentleman in Europe. Cassell and Company Ltd., London. Edwards J. 1996. London Zoo from old photographs 1852– 1914. Edwards, London. Fortescue J. 1912. Narrative of the visit to India of their majesties King George V and Queen Mary and of the Coronation Durbar held at Delhi 12th December 1911. Macmillan and Co., London. Rookmaaker LC. 1998. The rhinoceros in captivity: a list of 2439 rhinoceroses kept from Roman times to 1994. SPB Publishing, The Hague. Rookmaaker LC. 2004. Fragments on the history of the rhinoceros in Nepal. Pachyderm 37:73–79. Smythies O. 1961. Ten thousand miles on elephants. Seeley Service, London. Taylor A. 2004. ‘Pig-sticking princes’: royal hunting, moral outrage, and the republican opposition to animal abuse in nineteenth- and early twentieth-century Britain. History 89(293):30–48. 97 de Iongh et al. Some observations on the presence of one-horned rhinos in the bas reliefs of the Angkor Wat temple complex, Cambodia H.H. de Iongh,1* H.H.T. Prins,2 N. van Strien3 and L.G. Rookmaaker4 1 Centre for Environmental Science, POB 9518, 2300 RA Leiden, The Netherlands; email: [email protected]; *corresponding author 2 Wageningen University, Tropical Nature Conservation and Vertebrate Ecology Group, Campo Ma’an Project, Bornsesteeg 69, 6708 PD Wageningen, The Netherlands 3 S Asia Coordinator, International Rhino Foundation, Julianaweg 2, 3941 DM Doorn, The Netherlands; email: [email protected] 4 Rhino Resource Center, c/o IUCN Species Survival Programme, 219c Huntingdon Road, Cambridge CB3 0DL, UK; email: [email protected] During a visit to the Angkor Wat temple complex near Siem Reap in central Cambodia, 2–6 December 2004, the first two authors observed the presence of onehorned rhinos in the bas reliefs of the northern and southern galleries of the temple. The Angkor Wat temple complex (fig. 1) was built under the reign of the Khmer-Hindu ‘king god’ Suryavarman II, who reigned from 1113 to 1150 C.E. (Common Era) (Roveda 2003). In the North Gallery (west wing) in a bas relief representing ‘the battle of Devas and Assuras’ we identified a one-horned rhino mounted by one of the gods (fig. 2). In the North Gallery (east wing) in a bas relief representing ‘the victory of Vishnu over Bana’ we also identified a one-horned rhino in front of a war chariot with one of the gods in the driving seat, bow spanned, ready to shoot and with one foot on the chariot the other on the rhino (fig. 3). In the South Gallery (east wing) in the representation of ‘heavens and hell’ we identified another onehorned rhino, which is attacking the damned people who are in hell (fig. 4). We believe that all three of the one-horned rhinos on these bas reliefs most probably represent Javan Figure 1. The Angkor Wat temple. 98 Pachyderm No. 38 January–June 2005 Rhinos in bas relief of Angkor Wat rhinos (Rhinoceros sondaicus) and not Indian rhinos (R. unicornis). Both rhino species have similar skinfolds and a pattern of skin warts, although in the Javan rhino they are less pronounced. A clear distinction between the Indian and Javan rhino is the skinfold on the shoulder, which for the Javan rhino continues along the back and gives the frontal part of the back and neck a saddle-like appearance. This saddle is present in the rhinos of the North Gallery (west wing) (fig. 2) and the South Gallery (fig. 4). Also the shape of the Javan rhino’s body and head is different from that of the Indian rhino, the latter carrying its head higher, having a more concave back and forehead. Contrary to our interpretation are the large round warts on the rhino of the South Gallery and the skinfold on the shoulder (fig. 4), which resemble those of an Indian rhino. Also the skinfolds of the ‘Javan rhino’, particularly in the South Gallery (fig. 4) are not completely accurate, since the horizontal abdominal skinfold continues on the belly, which is not the case in R. sondaicus. During our investigation in Angkor Wat, we did not encounter any rhino resembling the two-horned Sumatran rhino (Dicerorhinus sumatrensis). This is not surprising. Although there are records of rhinos occurring in most parts of Cambodia except the northern regions, none of these refer to an animal with two horns. Even though the posterior horn of the Sumatran rhino is often small and difficult to notice when seen in the jungle, it has been asserted that only the onehorned Javan rhino has ever been known in Cambodia (Rookmaaker 1980; Foose and Strien 1997). It is remarkable that the rhinoceros, despite its impressive size and power, has always played a relatively minor role in Hindu art and lore in India. Although a large number of animals, even those confined to a limited range in northern India, were associated with the Hindu gods as vahanas or sacred mounts, the rhinoceros never reached this status in Hindu mythology and iconography (Bautze 1985). In Nepal, the kings are obliged to perform the Tarpan ceremony once in their lifetime, in which rhino blood libations are offered to Hindu gods. Rhino hide plays a role in the Shradda, an elaborate religious ceremony performed by both Hindus and Buddhists in Nepal to commemorate parents or grandparents on the anniversary of their deaths (Martin 1984:417; 1985). It is unusual, and apparently confined to Khmer art, to find the rhino employed as a vahana for one of Pachyderm No. 38 January–June 2005 Figure 2. Detail of a bas relief in the North Gallery (west wing) of Angkor Wat, showing a one-horned mounted by the god Agni. Figure 3. One-horned rhino shown in a bas relief in the North Gallery (east wing) at Angkor Wat. Figure 4. A rhinoceros without rider in the South Gallery (east wing) of Angkor Wat. 99 de Iongh et al. the gods in these reliefs. This explains why archaeologists have difficulty to decide which god is seen riding on the rhino in this instance. Lohuizen-de Leeuw (1955) suggested that it was Kartthikeya, the god of war, partly because one of his attributes is a sword, khadga in Sanskrit, which is also the name for the rhinoceros. Moens (1948) explained that it may have been Skanda, the god associated with forest fires. But here we become entangled in the complicated genealogies and hierarchies of the Hindu pantheon. It is best, therefore, to follow the convincing argument by Stönner (1925) that the presence of flames around the representation of the god suggests that he was Agni, the god of fire. In Hindu iconography, Agni is usually depicted with two heads and sometimes four arms riding a goat. According to the Agni Purana his attributes are the rosary, a jar of water, the javelin, and what looks like a garland of flames. It is generally accepted at present that the rhino here depicted is carrying the god Agni (Brentjes 1978; Roveda 1997, 2003). Although there is no evidence identifying the architects of the temple, Roveda (2003) states that it is likely that Divakarapandita, the Brahman under service to Suryavarman, contributed to the concept and planning. This Brahman priest, who came from India, was born in 1040 and died in 1120 (C.E.). One of a long line of illustrious Brahmans who served the Khmer kings, Divarkarapandita also served under Suryavarman’s predecessors—kings Jayavarman VI and Dharanindravarman I. He died 30 years before the construction of Angkor Wat was completed, but he was most likely the temple’s architect. We believe that this Brahman architect was likely to have influenced aspects of the design of the one-horned rhinos depicted on the bas reliefs, such as the larger Indian rhino-like warts on the skin. The Brahman priests however, used local Khmer artisans for final work on the designs, and this may explain the predominant similarities with one-horned Javan rhinos in the bas reliefs of the Northern Gallery. The Javan rhino is known to have occurred in Cambodia until recent times (Rookmaker 1983). Therefore, it is likely that 100 local knowledge of the Javan rhino influenced the design and execution of these particular bas reliefs. References Bautze J. 1985. The problem of the Khadga (Rhinoceros unicornis) in the light of archaeological finds and art. In: Schotsmans J and Taddei M, (eds.), South Asian archaeology 1983: Papers from the Seventh International Conference of the Association of South Asian Archaeologists in Western Europe, held in the Musées Royaux d’Art et d’Histoire, Brussels, vol. 1. Instituto Universitario Orientale, Dipartimento di Studi Asiatici (Series Minor 23), Naples. p 405–433. Brentjes B. 1978. Die Nashörner in den alten orientalischen und afrikanischen Kulturen. Säugetierkundliche Mitteilungen 26:150–160. Foose TJ, Strien N van. 1997. Asian rhinos: status survey and conservation action plan, new edition. IUCN/SSC Asian Rhino Specialist Group, Gland, Switzerland. 104 p. Lohuizen-de Leeuw JE van. 1955. The Dikpalakas in ancient Java. Bijdragen tot de Taal-, Land- en Volkenkunde 111:356–384. Martin EB. 1984. They’re killing off the rhino. National Geographic 165(3):404–422, figs. 1–16. Martin EB. 1985. Religion, royalty and rhino conservation in Nepal. Oryx 19(1):11–16. Moens JL. 1948. De eenhoorn van Skanda. Tijdschrift voor Indische Taal-, Land- en Volkenkunde 82:347–361. Rookmaaker LC. 1980. The distribution of the rhinoceros in eastern India, Bangladesh, China and the Indo-Chinese region. Zoologische Anzeiger 205:253–268. Rookmaker LC. 1983. Bibliography of the rhinoceros: an analysis of the literature on the recent rhinoceroses in culture, history and biology. Balkema, Rotterdam. Roveda V. 1997. Khmer mythology. Thames and Hudson, London. Roveda V. 2003. Sacred Angkor: the carved reliefs of Angkor Wat. River Books, Bangkok. 290 p. Stönner H. 1925. Erklärung des Nashornreiters auf den Reliefs von Angkor-Vat. Artibus Asiae 1(2):12–130. Pachyderm No. 38 January–June 2005 Mineral prospecting in the Selous Game Reserve, Tanzania OPINION Mineral prospecting in the Selous Game Reserve and its dangers to rhino conservation Rolf D. Baldus PO Box 1519, Dar es Salaam, Tanzania; email: [email protected] Abstract The Tanzanian government has granted prospecting licences for precious stones within the black rhino range of the Selous Game Reserve. There is an international agreement that there should be no mining in World Heritage Sites like the Selous. Mining in other protected areas in Tanzania has resulted in an uncontrolled influx of illegal miners, associated widespread lawlessness, and serious environmental destruction. Twenty years ago oil prospecting and explorations for a hydropower scheme at Stiegler’s Gorge contributed to the near extinction of rhinos in the Selous. Together with a large dam presently being planned across the Ruvu River at the north-eastern tip of the Selous rhino range, the possibility of mining poses a major threat to the recovering but still fragile rhino population. Résumé Le gouvernement tanzanien a accordé des licences de prospection de pierres précieuses dans l’aire de répartition des rhinos noirs dans la Réserve de Faune de Selous. Il existe un accord international disant qu’il ne devrait y avoir aucun forage dans des Sites du Patrimoine mondial tels que le Selous. Les forages dans d’autres aires protégées tanzaniennes ont abouti à l’afflux non contrôlé de prospecteurs illégaux, à l’expansion de l’état de non-droit qui leur est associé, et à de graves destructions environnementales. Il y a vingt ans, les prospections pétrolières et les explorations pour un programme hydroélectrique dans la Gorge de Stiegler ont contribué à la quasi-extinction des rhinos dans le Selous. S’ajoutant au vaste barrage qui est actuellement prévu en travers de la rivière Ruvu, à la pointe nord-est de l’aire des rhinos au Selous, la possibilité de prospection minière pose une menace majeure à cette population en voie de reconstitution, mais encore fragile. Granting of prospecting licences for precious stones Mining has developed in recent years as an important economic sector in Tanzania, one of the world’s poorest countries. Minerals now constitute a major export, and economic and political pressures are increasing to extend prospecting and mining into the vast network of protected areas. Recently the Ministry of Energy and Minerals granted licences to prospect for precious stones in Pachyderm No. 38 January–June 2005 various parts of the Selous Game Reserve. The reserve is one of the largest (close to 50,000 km2) protected areas in the world, the oldest in Africa (started in 1896) and it has been a World Heritage Site since 1982. Thirteen prospecting licences were allocated to three companies between 26 October 2002 and 5 June 2003 according to a letter of the Ministry of Energy and Minerals of 17 February 2004 (Tanzania 2004). The minister for Natural Resources and Tourism confirmed this in the press (Mbiro 2004) after the issue 101 Baldus TANZANIA Matambwe Msolwa Kingupira Miguruwe Ilonga Liwale 0 20 40 60 km Legend sector office Likuyu Seka TAZARA railway river Kalulu photographic tourist block hunting block Figure 1. Selous Game Reserve, Tanzania. 102 Pachyderm No. 38 January–June 2005 Mineral prospecting in the Selous Game Reserve, Tanzania was made public through an article in a local newspaper in Dar es Salaam (Baldus and Ngoti 2004). Further applications for prospecting have subsequently been received. No environmental impact assessment of the planned prospecting activities has been made. Dangers for the Selous rhino population All 13 prospecting concession areas granted are within the rhino range or close to it. Similar projects, deemed to have been environmentally destructive, took place in the 1980s. Efforts to build a dam and a hydroelectric scheme brought up to 2000 workers to Stiegler’s Gorge at the Rufiji River in the north-west of the reserve. The many rhinos in that area disappeared at the same time (Stephenson 1986). In the early 1980s while the search for oil was ongoing, thousands of kilometres of straight-cut lines were bulldozed through the Selous. The lines opened up formerly impenetrable thickets, providing access for poaching gangs of up to 60 people. Both projects greatly contributed to the near-complete loss of the rhino population, which at the time was estimated to be around 3000 and also facilitated the poaching of more than 50,000 elephants (Stephenson 1986). Alpers (2004) estimates that the rhino population in the Selous is slowly increasing at present. The remnant population in the northern (tourist) sector in the Selous has been under intensive protection and observation by a special project for several years, financed by the European Union and managed by the Wildlife Division together with the Selous Rhino Trust (Morgan-Davies 2001). The population is known to be breeding, and 20 animals are individually known. They are expanding their range and are increasingly seen by tourists and scouts. Five additional separate subpopulations south of the Rufiji River are also known and receive protection through the anti-poaching units of the reserve. No signs of rhino poaching have been found in the last 15 years though smallscale elephant poaching is going on within the rhino range (pers. comm. with Game Reserve management). Management in the Selous has greatly improved wildlife security in recent years (Baldus et al. 2003), but it would be difficult to cope with a large influx of people and the activities connected with mining precious stones, in particular if they take place illegally. Pachyderm No. 38 January–June 2005 Whereas mining by large companies is supervised, their activities are normally followed by those of illegal diggers. Artisanal mining might contribute to poverty reduction, but it is also related to a high degree of lawlessness, which is almost impossible to control in large and remote protected areas. This is illustrated by experiences of mining in the Eastern Arc Mountains, in particular in the Amani Nature Reserve in northern Tanzania (Nyiti 2004; Burgess et al. 2005). Similar environmental destruction has been caused by illegal mining in the Kizigo Game Reserve in north-western Tanzania (John Mbwiliza, pers. comm.).. In recent years uranium deposits in the Selous have been surveyed using a helicopter. There is concern that such activity may be followed by more environmentally damaging ground exploration and thereafter possibly by exploitation. But even if artisanal mining is prohibited in the Selous, doubts remain whether companies that have been granted prospecting licences will follow internationally agreed standards. They are not members of the Tanzanian Chamber of Mining, which regards the question of mining inside protected areas as sensitive but so far does not have a clear policy on the matter. Legal background Tanzania is internationally respected for its wildlife conservation policies. Its extensive network of protected areas covers approximately 25% of the total land area and consists of 13 national parks, 31 game reserves and the Ngorongoro Conservation Area. Wildlife management areas are an additional category of conservation range under community management. Protected areas in Tanzania are set aside and managed under various authorities. Tanzania National Parks (TANAPA) is the parastatal organization empowered to manage the national parks. The game reserves including the Selous are under the Wildlife Division, which is part of the Ministry of Natural Resources and Tourism. The Ngorongoro Conservation Area Authority is a separate parastatal. There are mineral deposits in many of these areas. The Ministry of Energy and Minerals will issue prospecting and mining licences regardless of whether the area is protected or not, but makes it quite clear that the final responsibility for exploration and mining in 103 Spike Williamson/Beho Beho Lodge Baldus A Selous rhino, photographed in 2004 near one of the planned prospecting sites. national parks and game reserves lies with the respective conservation authority. The Wildlife Conservation Act of 1974 does not contain any provisions on mining in game reserves or other areas under the Wildlife Division’s jurisdiction. However, the Act prohibits entrance without written permission from the director of Wildlife. This indirectly controls mining in game reserves. National Parks ordinances contain provisions that allow mining in national parks under certain conditions. TANAPA must grant permission to enter any national park and can thus control all activities on their estate. According to TANAPA’ s director general (DG), its Board of Trustees has ruled not to allow mining in any of the national parks in the country (Gerald Bigurube, TANAPA DG, pers. comm.) But Selous is not a national park, and no such ruling exists for national reserves. In the first meeting of the permanent subcommittee on wildlife conservation and management under the East African Community Secretariat, the three partner states agreed that the exploration and exploitation of mineral resources in protected areas should be prohibited (EACS 2004). For cases in which such resources were critical to the national economies, restrictive conditions for mining were formulated. 104 The minister of Natural Resources and Tourism, who has the responsibility for conservation areas in the country, was quoted in the local press as defending the granting of prospecting licences in the Selous since they establish the existence of minerals but do not necessarily allow their subsequent extraction (Mbiro 2004). Mining activity in World Heritage Sites IUCN, the World Conservation Union, through its World Heritage Committee recently initiated a dialogue with the International Council on Mining and Metals—a body representing 15 of the world’s largest mining and metal-producing companies. In August 2003 the council announced that its corporate members committed themselves not to explore or mine in any World Heritage Site and to take all possible steps to ensure that any mining operations present are compatible with the outstanding universal values of such sites (ICMM 2003). After this announcement the IUCN director general commented that a new threshold for corporate responsibility had been created and, notwithstanding several outstanding issues, the decision was an important milestone (ICMM 2003). Pachyderm No. 38 January–June 2005 Mineral prospecting in the Selous Game Reserve, Tanzania Discussion Disclaimer The political responsibility for prospecting and mining in any protected area of Tanzania ultimately lies with the Ministry of Natural Resources and Tourism. An anomalous situation now exists—one conservation agency of this ministry has banned mineral prospecting (in national parks only) while another appears about to permit it. Based on past experience there is a high probability that mining activities in the Selous would have negative consequences for the area’s environment and its biodiversity, including endangering the recovering elephant and rhino populations. It is difficult to understand why the Tanzanian authorities should allow mineral prospecting in protected areas, as this would seem to imply permitting future extraction, especially as leading players in the world’s mining industry have agreed that World Heritage Sites must be strictly exempted from mining activities of any kind. There are still many unprotected areas in Tanzania containing minerals that are not yet mined. It might make sense to prohibit mineral prospecting and mining in all conservation areas until mining can in future meet far more stringent and enforceable environmental standards. It is of interest to note that De Beers of South Africa prospected much of the Selous for precious stones in the latter half of the 1950s; they found nothing worth following up (B. Nicholson, email 19 March 2005). Unfortunately mining is not the only imminent danger for the Selous rhino population. The Tanzanian government plans to construct a large dam across the Ruvu River at the north-eastern edge of the Selous Game Reserve to supply Dar es Salaam with water. The dam will not only destroy several hundred square kilometres of dry-season grazing land that is indispensable for the wildlife of the northern Selous, but will also flood parts of the rhino range. Like 25 years ago, this project will once again introduce a large workforce close to these specially protected rhinos. This article expresses the views of the author only and not necessarily those of Pachyderm or the organizations he works for. Pachyderm No. 38 January–June 2005 References Alpers F. 2004. Selous Black Rhinoceros Protection Project. Internal paper. Wildlife Division, European Union and Selous Rhino Trust, Dar es Salaam. Baldus RD, Ngoti PM. 2004. Mining, conservation put Selous on the cross road. Daily News 20 July 2004, Dar es Salaam. Baldus RD, Kibonde B, Siege L. 2003. Seeking conservation partnerships in the Selous Game Reserve, Tanzania. Parks 13(1):50–61. Burgess N, Kilahama F, Nderumaki M, Kahemela A, Sawe C. 2005. Mining in the Eastern Arc Mountains: the situation in early October 2004. Oryx 39(1):13. [EACS] East African Community Secretariat. 2004. Report of the 1st meeting of the Permanent Sub-committee on Wildlife Conservation and Management, Arusha. Unpublished paper. [ICMM] International Council on Mining and Metals. 2003. Landmark ‘no-go’ pledge from leading mining companies. Press release, London, 20 August 2003. Mbiro M. 2004. Government allows mining prospecting in Selous. Daily News 22 July 2004, Dar es Salaam. Tanzania. Ministry of Energy and Minerals. Allocation of mining/prospecting plots inside the Selous Game Reserve. Letter. Ref. no. CAB.233/433/01/22 dated 1 July 2004 to Daily News, Dar es Salaam. Morgan-Davies M. 2001. Survey and conservation status of five black rhino (Diceros bicornis minor) populations in the Selous Game Reserve, Tanzania, 1997–1999. Pachyderm 31:21–35. Nyiti P. 2004. Gold rush threatens the ecosystem. Daily News 9 April 2004, Dar es Salaam. Stephenson I. 1986. The Selous Game Reserve in crisis. Frankfurt Zoological Society. Unpublished paper. 105 Mubalama and Mbula FIELD NOTE Less room for a small population of elephants in severely encroached Mikeno massif, southern Virunga National Park, Democratic Republic of Congo Leonard Mubalama1* and Déo Mbula2 1 Monitoring the Illegal Killing of Elephants (MIKE/CITES) Programme, PO Box 852, Bukavu, Democratic Republic of Congo; email: [email protected]; *corresponding author 2 Institut Congolais pour la Conservation de la Nature (ICCN/PNVi), BP 660, Goma, Democratic Republic of Congo; email: [email protected] Introduction Expanding human activity and incursions in the Mikeno forest zone in Virunga National Park (VNP) need to be addressed with an elephant conservation strategy in mind, even though poaching is the most urgent shortterm threat . The park recently experienced changes to its natural habitat that have been unprecedented in historic times. Habitat fragmentation is widespread and profound and has taken a heavy toll on wildlife, including a small population of elephants there. In a series of anti-poaching operations, 128 weapons were seized, 522 snares dismantled, 4 people arrested, 2 locally made arms recovered, and found was a mound of small mammal carcasses too numerous to count (ICCN 2004). The region remains at high risk, and Mikeno’s extraordinary biodiversity will be lost if uncontrolled hunting, deforestation and habitat degradation continue. War has indeed opened up the Mikeno sector to unprecedented exploitation of natural resources. Even when the conflict ends and security recovers, these threats will continue to grow. This paper recommends mechanisms to protect the critically endangered habitat and wildlife of Mikeno massif. The rape of Mikeno Reports of extensive habitat destruction and land conversion by pastoralists were received in Goma (fig. 1) 106 in June 2004 when approximately 6000 people accompanied by Rwandan military personnel were said to have moved into the Mikeno sector of VNP, cutting down and rapidly destroying large expanses of bamboo and alpine forests (FZS et al. 2004). Large herds of livestock were reportedly introduced into the park. Sources pointed to the presence of large numbers of armed, uniformed personnel, apparently protecting livestock, and of people chopping down trees (FZS et al. 2004). In response to these reports, an aerial reconnaissance was undertaken on 12 June 2004, through the European Commission’s Development Office (ECDO) in Goma, Democratic Republic of Congo (DRCongo). The area was flown over twice at an altitude of about 1000 m above ground level. The damage to the natural vegetation was extensive, all woody vegetation having been cut and burned. The situation was desperate; habitat destruction occurred at times at a rate of up to 2 km2 per day (the whole of the gorilla sector is less than 250 km2). Approximately 15 km2 of the park’s habitat was destroyed between 28 May and 12 June 2004 (ECDO 2004). The former local chief of Kibumba area (fig. 1) benefited directly from this illegal land use, receiving Rwanda francs 5000 (USD 8) per hectare of park land cleared while workers employed to carry out this large-scale clearing received Rwanda francs 500 (USD 0.80) per person per day from the Rwandan side (FZS et al. 2004). Pachyderm No. 38 January–June 2005 Elephant population in Virunga National Park, DRCongo Batoui Kara II Kalirangiri Nyamulagira VIRUNGA NATIONAL PARK (DR CONGO) Tongo Rigo Baringo Akenge Jomba Rumangabo Bukima Sabinyo Kakomero Batoui Nyamulagira Mikeno Visoke RigoKibumba Karisimbi Nyiragongo VOLCANOES NATIONAL PARK (RWANDA) Kilgi Mugunga Kibumba MGAHINGA NATIONAL PARK (UGANDA) Kalengera Mishari Mikeno Legend N approximately 15 km2 Mwaro corridor area Mikeno encroached area international border mountain summit patrol post road network park limit (southern sector) Goma 16 0 32 64 km Figure 1. The Mikeno encroached area. The sector of the forest cleared is important habitat that harbours a large variety of species. The loss of 15 km2 in this part of the park was accompanied by the loss of many valuable services that forests provide such as regulating hydrological flows and sequestering carbon, and by loss of the biodiversity they contain including the mountain gorilla (Gorilla beringei beringei), the endemic golden monkey (Cercopithecus mitis kandti), l’Hoest’s monkey (Cercopithecus l’hoesti), buffalo (Syncerus caffer), duiker (Cephalophus spp.) and elephant (Loxodonta africana ssp.). Habitat destruction On small open fires under the cooking pots of peasant families, thousands of tonnes of wood—a product of park encroachment—go up in smoke every day. This encroachment marks the significant changes that have occurred in the surrounding area of the recently deforested wildlife conservation habitat. Five factors are catalysing the destruction in Mikeno sector: 1) clearing for cultivation; 2) grazing and dry-season burning stimulating new growth of grasses for grazing cattle at a time when other fodder is scarce; 3) attempting to destroy the refuge of the undesirable interahamwe and to improve visibility in Pachyderm No. 38 January–June 2005 tall grass areas suspected to be their hide-out; 4) collecting firewood; and 5) increasing the supply of tall thatching grass. Over the years, conservation in the Mikeno sector has faced problems, most if not all of which hinge on high population pressure and demand for livelihood needs (Sikubwabo and Mushenzi 1996). The loss of any amount of forest land to agricultural encroachment, the illegal use of forest products, and poachers and habitat destruction have left just a small percentage of the originally abundant resources, which must now be actively controlled to ensure their continued existence. Elephants and gorillas are dependent on forest habitat, and once the forest is converted to agriculture it rarely reverts to natural vegetation, unless it is left for hundreds of years. Deforestation means less room for elephants (Barnes 1990), and degradation of the forest is ongoing in an area where the carrying capacity is estimated to be 120 elephants, calculating an average of 0.5 elephant per km2 (based on Pfeffer 1989). Even where the forest is not destroyed, it may be effectively lost as wildlife habitat if disturbance is severe. The further spread of villages and agriculture will inevitably increase competition between humans and elephants for land in the Mikeno forest zone, now well known as a hideout for heavily armed Rwandan soldiers. 107 Mubalama and Mbula Elephant corridor, genetic diversity The sensitive forest corridor at Mwaro that joins Mikeno’s vulnerable habitat with the rest of the park has also been partially affected (fig. 1) by providing spare land with water and wood in one of the most densely populated regions in the country. The Mikeno sector contains about 75 elephants (Blanc et al. 2003). Evidence suggests that isolated and small populations are less genetically diverse than contiguous large ones (Sheppard 1975; Berry 1977). Although Mikeno habitat does not yet appear to be in such dire straits and no factual basis for this statement is so far provided, the question should be: ‘Is the lack of variation a danger to a small, isolated population?’ The answer among conservationists seems to be ‘yes’. Elephants and buffaloes in particular use this 800m corridor along their migration route, and its potential rapid disappearance presents a real threat to the genetic diversity of local fauna by preventing gene flow from the Mikeno sector to the other areas of the park such as the Nyamulagira sector, and indeed to other countries (fig. 1). Recommendations Genetic diversity Conservationists should certainly be aware and make others aware of the possible genetic dangers of small population sizes in such isolated contexts. However, considering the lack of evidence on the subject, some have been, perhaps, too emphatic about these hazards. Environmental and demographic factors, not genetic ones, are, we suggest, the danger to such small and declining populations. Their conservation is necessary to ensure that the genetic and behavioural diversity of species is preserved. Therefore, there is critical need to carry out a further study on the pool of genetic diversity of the remaining elephant population in Mikeno to find out whether it can survive large-scale habitat destruction exceeding the limits of development plasticity. Human population and encroachment In the long term, human population growth and the expansion of settlement and agriculture will result in the loss of habitat (Mubalama 1995). The human population around the Virunga volcanoes region is increasing 108 at 3% per annum (Nsabimana 1978), which means it will double in just 24 years. Population distribution is uneven, with peak densities found on the rich volcanic soils around the Virunga volcanoes. With 95% of the population living on agriculture (Nsabimana 1978) immense demands are obviously placed on the land, demands that will become ever more pressing as time goes on. Somehow we must find a way between total appropriation and total protection so that, to the benefit of all, we can achieve some balance between humans and wildlife instead of open conflict. If not well monitored and protected, this small population may soon be engulfed by the ever-growing human population. Poaching, law enforcement and political will The park area, lying on the border between Rwanda and DRCongo, is considered a high-security zone used by poachers and militias. Although certain areas of the park are extremely dangerous, guards from the Institut Congolais pour la Conservation de la Nature are patrolling the international borders to ensure around-the-clock protection of key species. The guards generally have much less sophisticated weapons than the well-equipped militias. In addition, ammunition is limited. If the donor community is rallied to provide the guards with better equipment and to top up their salaries, we hope they will be encouraged to control ivory poaching. However, to deter poaching in the field, prevent the illicit transport of ivory within concerned countries, and stop the smuggling of ivory within and outside the Virunga volcanoes region requires determined government action. At the time of submitting this field note, we heard from staff in Goma that 300 kg of ivory were going in to business through the black market around the park. To curb such trafficking, we urge cooperation and coordination between national CITES authorities and wildlife law-enforcement agencies including customs. Lack of political support to improve coordination of wildlife enforcement remains a persistent problem. It is a problem that can be dealt with through the Lusaka Agreement Task Force (LATF), yet only Uganda among the three concerned countries has so far ratified it. Therefore, we recommend that DRCongo and Rwanda also ratify the LATF to improve coordination of law enforcement. Pachyderm No. 38 January–June 2005 Elephant population in Virunga National Park, DRCongo Everybody’s problem References The effects of the recent encroachment upon conservation in the area are still fresh and continue to be a great loss. The cost of the recovery programme will be enormous, and even so, not everything lost will be recovered. If the relevant authorities are made aware of the capital based on natural resources that is being lost in terms of individual wildlife species and related revenue, it should be an eye-opener to them, and to prosper, they should protect the region zealously. According to MacKinnon et al. (1986), there is no foolproof management prescription to protect parks during war. However, public support is crucial and an international link invaluable. If elephants and their habitat are to be protected, efforts must also aim at bringing all surrounding Mikeno local people into a clear-cut sensitization programme. Even so, the international demand for illegal ivory traffic to Japan will still result in more forest elephants being poached (Nishihara 2003). Wardens and park guards alone cannot stop it. Other government departments must be involved, such as the general police force, the criminal investigations department, the judiciary, customs and excise, the intelligence services and the general public (Mubalama and Mushenzi 2004). The International Gorilla Conservation Programme (IGCP) has given us a valuable message. Its programme straddles the three countries in which Mikeno falls— DRCongo, Rwanda and Uganda—and it works closely with all the surrounding people. IGCP raised the alarm when Rwandan military and pastoralists invaded Mikeno, pointing out the detrimental effect they were having on the habitat and the elephants. And with help from EU and others action was taken to remove the invaders. International efforts must be made to protect this relict and isolated population. To promote its long-term viability, and by implication its habitat, remote sensing can be used to monitor changes in the ecosystem and the degree of fragmentation, from which to track the changing extent and intensity of key threats. The degree of human disturbance should be monitored across all areas of the forest, with a continual check on the human effect on wildlife, to inform protection programmes as efficiently as possible. Barnes RFW. 1990. Deforestation trends in tropical Africa. African Journal of Ecology 28:161–173. Berry RJ. 1977. Inheritance and natural selection. Collins, London. Blanc JJ, Thouless, CR, Hart JA, Dublin HT, Douglas-Hamilton I, Craig CG, Barnes RF W. 2003. African elephants status report 2002: an update from the African Elephant Database. IUCN/SSC African Elephant Specialist Group, Gland, Switzerland. [ECDO] European Commission’s Development Office. 2004. Report of an aerial reconnaissance flight revealing the catastrophic destruction of Virunga National Park’s Mikeno gorilla sector. Site report, 13 June 2004. FZS, IGCP, WWF, WCS, ZSL. 2004. A preliminary report following a meeting between Frankfurt Zoological Society, International Gorilla Conservation Programme, World Wide Fund for Nature, Wildlife Conservation Society and Zoological Society of London concerning deforestation in the Mikeno sector. 3 p. [ICCN] Institut Congolais pour la Conservation de la Nature. 2004. Rapport technique. UNF/UNESCO, Goma, DRCongo. MacKinnon J, MacKinnon K, Child G, Thorsell J. 1986. Managing protected areas in the tropics. IUCN, Gland, Switzerland. Mubalama LK. 1995. The Virunga National Park grappling with the problem of population pressure: a case study of human ecology. Wildlife and Nature 11(2):52–56. Mubalama L, Mushenzi N. 2004. Monitoring law enforcement and illegal activities in the northern sector of the Parc National des Virunga, Democratic Republic of Congo. Pachyderm 36:16–29. Nishihara T. 2003. Elephant poaching and ivory trafficking in African tropical forests with special reference to the Republic of Congo. Pachyderm 34:66–74. Nsabimana D. 1978. La conservation de la nature au Rwanda. Address to the IUCN General Assembly, Ashkhabad, September 1978. Pfeffer P. 1989. Vie et mort d’un géant : l’éléphant d’Afrique. L’Odyssée/Flammarion, France. Sheppard PM. 1975. Natural selection and heredity. Hutchinson University Library, London. Sikubwabo C, Mushenzi N. 1996. Mountain gorillas of Mikeno, Zaïre: an explosive situation. Unpublished report. Pachyderm No. 38 January–June 2005 109 GUIDELINES TO CONTRIBUTORS Aim and scope Pachyderm publishes papers and notes concerning all aspects of the African elephant, the African rhino and the Asian rhino with a focus on the conservation and management of these species in the wild. At the same time, the journal is a platform for disseminating information concerning the activities of the African Elephant, the African Rhino, and the Asian Rhino Specialist Groups of the IUCN Species Survival Commission (SSC). Submission of manuscripts Submit manuscripts electronically by email. Alternatively, submit a hard copy and floppy disk or CD by mail. Email contributions should be sent to: [email protected] with copy to: [email protected] The Editor, Pachyderm IUCN/SSC AfESG PO Box 68200, GPO 00200 Nairobi, Kenya tel: +254 20 576461; fax: +254 20 570385 Preparation of manuscripts Manuscripts are accepted in both English and French languages. Where possible, the abstract should be provided in both languages. Title and authors: The title should contain as many of the key words as possible but should not be more than 25 words long. Follow with the name(s) of the author(s) with insitutional affiliation and full postal and email address(es). Indicate the corresponding author, to whom proofs and editorial comments will be sent; give post, fax and email addresses for the corresponding author. 110 Research papers: Should be not more than 5000 words and be structured as follows: 1) Title (as above), 2) Abstract of not more than 250 words (informative type, outlining information from the Introduction, Materials and methods, Results, Discussion, but not detailed results), 3) additional key words (if any), not appearing in the title, 4) Introduction, 5) Materials and methods, 6) Results, 7) Discussion, 8) Conclusions if appropriate, 9) Acknowledgements (optional, brief), 10) References, 11) Tables, 12) Figure and photo captions, 13) Figures and photos. Papers may be reports of original biology research or they may focus more on the socio-economic aspects of conservation, including market surveys. Preferably provide figures and maps in their original form, for example, Excel files, maps as eps or tif files (17 x 15 cm, 600 dpi), when submitting in electronic form. Indicate clearly the author or source of figures, maps and photographs. Field notes: The journal welcomes notes from the field. They may contain figures and tables but should be brief. Book reviews: Pachyderm invites reviews of newly published books, which should be no more than 1500 words long. Letters to the editor: Letters are welcome that comment on articles published in Pachyderm or on any other issue relating to elephant and rhino conservation in the wild. Journal conventions Nomenclature Use common names of animals and plants, giving scientific names in italics on first mention. Pachyderm No. 38 January–June 2005 Guidelines Use an ‘s’ for the plural form for animals: rhinos, elephants. In the reference list, cite publications as in the following examples. List in alphabetical order. Write out journal titles in full. Spelling Use British spelling, following the latest (10th) edition of the Concise Oxford Dictionary, using ‘z’ instead of ‘s’ in words like ‘recognize’, ‘organization’, ‘immobilized’; but ‘analyse’, ‘paralyse’. Numbers Use SI units for measurement (m, km, g, ha, h) with a space between the numeral and the unit of measurement. Give measurements in figures, for example 12 mm, 1 km, 3 ha, except at the beginning of a sentence. Spell out numbers under 10 if not a unit of measurement unless the number is part of a series containing numbers 10 or over, for example: 14 adult males, 23 adult females and 3 juveniles. In the text, write four-digit numbers without a comma; use a comma as the separator for figures five digits or more: 1750, 11,750. The separator will be a full stop in French papers. References Use the author-year method of citing and listing references. In the text, cite two authors: ‘(X and Y 1999)’ or ‘X and Y (1999)’; cite more than two authors ‘(X et al. 1996)’ or ‘X et al. (1996)’. Note that there is no comma between the author(s) and the year. Pachyderm No. 38 January–June 2005 Adams JX. 1995b. Seizures and prosecutions. TRAFFIC Bulletin 15(3):118. Dobson AP, May RM. 1986. Disease and conservation. In: ME Soulé, ed., Conservation biology: the science of scarcity and diversity. Sinauer Associates, Sunderland, MA. p. 123–142. Struhsaker TT, Lwanga JS, Kasenene JM. 1996. Elephants, selective logging and forest regeneration in the Kibale Forest, Uganda. Journal of Tropical Ecology 12:45–64. Sukumar R. 1989. The Asian elephant: ecology and management. Cambridge Studies in Applied Ecology and Resource Management. Cambridge University Press, Cambridge. Cite unpublished reports as follows: Tchamba MN. 1996. Elephants and their interactions with people and vegetation in the Waza–Logone region, Cameroon. PhD thesis, University of Utrecht, The Netherlands. 142 p. Unpublished. Woodford MH. 2001. [Title]. [Journal or publisher]. Forthcoming. [if publication date is known] Woodford MH. [Title]. [Journal or publisher]. In press. [if publication date is not known] Government reports, reports to wildlife departments, MSc theses, PhD theses, etc. are to be noted as unpublished. Not accepted as references are papers in preparation or submitted but not yet accepted. ‘Pers. comm.’ accompanied by the date and name of the person is cited in the text but not given in the reference list. 111