journal of the african elephant, african rhino and asian rhino

Journal of the African Elephant, African Rhinoand Asian Rhino Specialist Groups

July – December 2012 No. 52

EditorsDali Mwagore and Helen van Houten

Section EditorsDeborah Gibson–African elephantsKees Rookmaaker–African and Asian rhinos

Editorial BoardJulian BlancHolly T. DublinRichard EmslieMike KnightEsmond MartinBenson Okita-OumaRobert OlivierDiane SkinnerBibhab K. TalukdarLucy Vigne

Design and layoutDali Mwagore

IllustrationsPhilip Miyare

Address all correspondence,including enquiries about subscription, to

The Editor, PachydermPO Box 68200, 00200Nairobi, Kenyatel: +254 20 249 3561/65fax: +254 20 2493570email: [email protected]: http://african-elephant.org http://pachydermjournal.org

Reproduction of this publication for educational or other non-commercial purposes is authorized without written permission from the copyright holder provided the source is fully acknowledged.

ISSN 1026 2881

1 Chair reports / Rapports des Présidents

1 African Elephant Specialist Group report/ Rapport du Groupe des Spécialistes des Eléphants d’Afrique

Holly T. Dublin

7 African Rhino Specialist Group report/Rapport du Groupe des Spécialistes des Rhinocéros d’Afrique

Mike Knight

20 Asian Rhino Specialist Group report/Rapport du Groupe des Spécialistes des Rhinocéros d’Asie

Bibhab K. Talukdar

23 Research

23 Functional relationship between crop raiding by the savannah elephant and habitat variables of the Red Volta Valley in north-eastern Ghana

Patrick Adjewodah, William Oduro and Alex Asase

36 Indentification et caractérisation des formations végétales exploitées par l’éléphant Loxodonta africana dans la Réserve de Biosphère de la Pendjari au Nord-Ouest de la République du Bénin

Tehou C. Aristide, Kossou Eric, Mensah G. Apolinaire, Houinato Marcel and Sinsin Augustin Brice

49 Optimizing the habitat of the Javan rhinoceros (Rhinoceros sondaicus) in Ujung Kulon National Park by reducing the invasive palm Arenga obtusifolia

Adhi R.S. Hariyadi, Agus Priambudi, Ridwan Setiawan, Daryan, Hendra Purnama and

Asep Yayus

S p e c i e SS u r v i v a l

c o m m i S S i o n

Cover: African bull elephant in Ngorongoro, TanzaniaCredit: ©Sian Brown, Messerli Foundation


Journal of the African Elephant,

African Rhino and

Asian Rhino Specialist Groups

55 Demand for forest elephant ivory in Japan Tomoaki Nishihara

66 Dispersal and social behaviour of the three adult female white rhinos at Ziwa Rhino Sanctuary in the immediate period before, during and after calving

Felix J. Patton, Petra E. Campbell, Angie Genade, Robert Ayiko and Godfrey Lutalo

72 Management

72 Pilot study to validate PIKE-based inferences at site level Hugo Jachmann

88 Rhino/ Field notes

88 Notes on black rhino mortalities in North Luangwa National Park, Zambia Chansa Chomba, David Squarre and Harvey Kamboyi

91 New maps representing the historical and recent distribution of the African species of rhinoceros: Diceros bicornis, Ceratotherium simum and Ceratotherium cottoni

Kees Rookmaaker and Pierre-Olivier Antoine

97 Density-dependent effect affecting elephant seed-dispersed tree recruitment (Irvingia gabonensis) in Congo Forest

David Beaune, Loïc Bollache, Barbara Fruth, Gottfried Hohmann and François Bretagnolle

101 MIKE / ETIS updates

101 CITES-MIKE Update / Mise à jour de la CITES-MIKE Julian Blanc

106 Progress in implementing the Elephant Trade Information System (ETIS) /Avancement dans la mise en œuvre du Système d’Information sur le Trafic des Eléphants (ETIS)

Tom Milliken and Louisa Sangalakula

113 Guidelines for contributors

Views expressed in Pachyderm are those of the individual authors and do not necessarily reflect those of IUCN, the European Union, 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).

AcknowledgementsThe production of this issue of Pachyderm was possible through contributions from a number of organizations and individuals. In particular, we would like to thank the following:


Journal of the African Elephant,

African Rhino and

Asian Rhino Specialist Groups

EUROPEAN COMMISSION

Pachyderm No. 52 July–December 2012 1

African Elephant Specialist Group report

The past six months have seen a sharp increase in awareness of the current pressures on elephants from poaching and illegal ivory trade in many parts of their range, in large part as a result of IUCN’s joint reporting with MIKE and ETIS to the CITES Standing Committee in July 2012. High-profile press reports as well as statements from the United States State Department, IUCN’s resolution (see below) and initiatives by a variety of NGOs indicate that the issue is now getting attention at many levels. There is a significant amount of goodwill out there, and all indications are that people are gaining a deeper understanding of the dynamics of the problem. We are all reflecting on the failure of so many efforts to date and focusing on identifying what actions might result in positive outcomes. In all this, it is equally challenging to find a way for elephant conservationists to have impact in areas with which we have not previously had experience, such as consumer choice in China and law enforcement at border crossings, to name just a couple. It is increasingly clear that we need a much broader alliance, both within governments and among the NGO and intergovernmental organization community, to come up with innovative and effective solutions all along the illegal ivory supply chain. At the same time, we must continue to work on the ground to secure elephant habitat and corridors. These two strands of work are essential to secure the future for the African elephant.

African Elephant Specialist Group reportRapport du Groupe des Specialistes des Eléphants d’Afrique

Holly T. Dublin, Chair/Président

IUCN/SSC African Elephant Specialist Group, PO Box 68200 – 00200, Nairobi, Kenya email: [email protected]

Les six derniers mois ont vu une forte augmentation de la prise de conscience des pressions actuelles sur les éléphants émanant du braconnage et du commerce illégal de l’ivoire dans de nombreuses zones de leur habitat, en grande partie en raison des rapports conjoints de l’UICN avec MIKE et ETIS au Comité permanent de la CITES en juillet 2012. Des rapports à la une dans la presse ainsi que des déclarations du Département d’Etat Américain, la résolution de l’UICN (voir ci-dessous) et les initiatives prises par diverses ONG indiquent que la question attire maintenant l’attention à plusieurs niveaux. Il y a de la bonne volonté, et tout porte à croire que les gens comprennent mieux la dynamique du problème. Nous réfléchissons tous sur l’échec de tant d’efforts jusqu’à ce jour et nous nous concentrons sur l’identification des mesures qui pourraient aboutir à des résultats positifs. Dans tout cela, c’est aussi un défi de savoir comment les défenseurs des éléphants peuvent avoir un impact dans les domaines dans lesquels nous n’avons pas encore eu l’expérience, comme le choix des consommateurs en Chine et l’application de la loi aux frontières, pour n’en citer que quelques-uns. Il est de plus en plus clair que nous avons besoin d’une alliance beaucoup plus large, à la fois parmi les gouvernements, les ONG et la communauté des organisations intergouvernementales, pour proposer des solutions innovatrices et efficaces tout au long de la chaîne d’approvisionnement illégal de l’ivoire. En même temps, nous devons continuer à travailler sur le terrain pour sécuriser l’habitat et les corridors des éléphants. Ces deux axes de travail sont essentiels pour assurer l’avenir de l’éléphant d’Afrique.

CHAIR REPORTS

2 Pachyderm No. 52 July–December 2012

Dublin

IUCN’s 5th World Conservation CongressThe 5th session of the IUCN World Conservation Congress was held in Jeju, Republic of Korea, from 6 to 15 September 2012. Diane Skinner and I both attended the congress; Diane Skinner (AfESG programme officer) as an official member of the congress team. At the Species Pavilion we presented on elephants and poaching, and we participated in discussions of the wild meat trade and held a Knowledge Café on the African and Asian Elephant Database (AAED).

The most important outcome for elephants, however, was the passing of Resolution 025 ‘Conservation of African elephants’ (http://portals.iucn.org/docs/iucnpolicy/2012-resolutions/en/WCC-2012-Res-025-EN%20Conservation%20of%20African% 20elephants.pdf), that calls on IUCN to take a number of actions, most urgently to convene a high-level meeting on the growing concern with regard to elephant poaching and illegal ivory trade. AfESG has been working closely with the German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety, the lead sponsor of the original motion, to fine-tune the scope and aims of the proposed meeting. The intention is to raise awareness of poaching and the illegal ivory trade all along the supply chain—from source to destination—at the highest level within governments. In each country, different levels of government present barriers to effectively protecting elephants or reducing the illegal trade in ivory. In some cases, the defence forces or customs authorities may be problematic; in others, resources allocated to the wildlife management agency may be insufficient to effectively protect elephants in the field. It is therefore necessary to raise this issue with the highest level within each state, and to come together to determine the actions that each can take, individually and collectively, to stem the illegal ivory trade and to ensure that elephant populations are protected or recovering.

With the resolution now in place, IUCN now has the mandate to take on this work, but resources are tighter than ever. Diane and I have been working closely with the German government, the UK government, the offices of the IUCN president and director general and the IUCN Secretariat to raise the funds required to implement the resolution.

5ème Congrès mondial de l’UICN sur la ConservationLa 5ème session du Congrès mondial de l’UICN s’est tenue à Jeju en République de Corée, du 6 au 15 septembre 2012. Diane et moi avons toutes les deux participé au Congrès, Diane en tant que membre officiel de l’équipe du Congrès. Au Pavillon des Espèces, nous avons fait une présentation sur les éléphants et le braconnage, participé à des discussions sur le commerce de la viande de brousse et tenu un Café des connaissances sur la BDEAA.

Cependant, le résultat le plus important pour les éléphants a été l’adoption de la Résolution 025 « Conservation des éléphants d’Afrique » (http://portals.iucn.org/docs/iucnpolicy/2012-resolutions/en/WCC-2012-Res-025-EN%20Conservation%20of%20African% 20elephants.pdf), qui demande à l’UICN de prendre un certain nombre d’actions et de convoquer de façon urgente une réunion de haut niveau sur les préoccupations croissantes relatives au braconnage des éléphants et au commerce illégal de l’ivoire. Le GSEAf travaille en étroite collaboration avec le Ministère Fédéral allemand de l’Environnement, Conservation de la Nature et Sécurité Nucléaire, le principal commanditaire de la motion originale, pour peaufiner les objectifs de la réunion proposée. Le but est de faire la sensibilisation au plus haut niveau des gouvernements sur le braconnage et le commerce illégal de l’ivoire tout au long de la chaîne d’approvisionnement, de la source à la destination. Dans chaque pays, il y a différents niveaux du gouvernement qui présentent des barrières à la protection efficace des éléphants ou à la réduction du commerce illégal de l’ivoire. Dans certains cas, les forces de défense ou les autorités douanières peuvent être problématiques; dans d’autres, les ressources allouées à l’agence de gestion de la faune sont insuffisantes pour protéger efficacement les éléphants sur le terrain. Il faut donc soulever cette question au plus haut niveau de chaque Etat, et se réunir pour déterminer les actions que chacun peut prendre, individuellement et collectivement, pour enrayer le commerce illégal de l’ivoire et s’assurer que les populations d’éléphants sont protégées ou se rétablissent.

Avec la résolution maintenant en place, l’UICN a le mandat d’entreprendre ce travail, mais les ressources sont plus limitées que jamais. Diane et moi avons travaillé en étroite collaboration avec le gouvernement allemand, le gouvernement britannique, les bureaux du Président de l’UICN, le Directeur général et le Secrétariat de l’UICN



The African and Asian Elephant Database (AAED)Peter Mwangi (AfESG database officer) has continued to enter and analyse the surveys of elephant populations received since the publication of the last status report in 2007. Almost 150 submissions have been made, including many by data providers themselves. Peter is now working on developing simple how-to tools to assist data providers to upload their reports themselves to our new online system.

While we are still awaiting a couple of reports from surveys conducted in 2011, we have prepared a set of pooled estimates with a data cut-off date at the end of 2011. The numbers are currently under review by AfESG’s Data Review Working Group. As a general observation, there have, once again, been few repeat surveys, making it extremely difficult to infer trends. The amount of new data coming from West Africa has been little, but thankfully, the quality and the coverage of data for Central Africa have notably improved. Coverage in East and Southern Africa remains much the same, with several repeat surveys in important populations. However, a number of important methodological and seasonal adjustments in the underlying surveys make inferring trends in these two subregions still quite challenging.

While we know there is a high demand to see the updated estimates, the voluntary time available from our survey experts is more limited than ever in current economic times. Nonetheless, we do hope to have the estimates out as soon as possible. We are also aware that there is a desire to see detailed trend analyses. This is not possible with the current quality and frequency of data available but could perhaps be achieved by using models. We would be pleased to hear from anyone who has interest in taking this possibility forward, recognizing that AfESG currently has no resources for such a process.

Having digitized all the survey boundaries where data providers had not given us shapefiles, Peter has also begun work on the new range map. This is a time-consuming part of the process and we appeal to data providers if at all possible to send us the boundary shapefiles along with their much-appreciated survey reports.

pour réunir les fonds nécessaires pour mettre en œuvre la résolution.

La Base de Données sur les Eléphants d’Afrique et d’Asie (BDEAA)Peter continue à saisir et à analyser les études des populations d’éléphants reçues depuis la publication du dernier rapport de situation en 2007. Près de 150 observations ont été faites, y compris un grand nombre par des fournisseurs de données eux-mêmes. Peter travaille actuellement sur le développement de simples instruments de guide pratique pour aider les fournisseurs de données à télécharger leurs rapports eux-mêmes dans notre nouveau système en ligne.

Alors que nous attendons encore quelques rapports des recensements menés en 2011, nous avons préparé une série d’estimations groupées avec une date limite des données à la fin de 2011. Le Groupe de Travail chargé de l’examen des données du GSEAf est en train de passer ces chiffres en revue. D’une manière générale, il y a eu encore une fois très peu de recensements répétés et il est donc extrêmement difficile de dégager des tendances. Il y a eu très peu de nouvelles données d’Afrique de l’Ouest, mais, heureusement, il y a eu une amélioration notable dans la qualité et la couverture des données pour l’Afrique centrale. La couverture en Afrique orientale et australe reste la même, avec la répétition de plusieurs recensements sur les populations importantes. Cependant, à cause des ajustements méthodologiques et saisonniers importants effectués lors des recensements, il est difficile de bien comprendre les tendances dans ces deux sous-régions. Alors que nous savons qu’il y a une forte demande de voir les prévisions actualisées, le temps dont disposent nos experts pour le volontariat est plus limité que jamais dans la situation économique actuelle. Néanmoins, nous espérons avoir des estimations dès que possible. Nous sommes également conscients qu’il y a un désir de voir des analyses détaillées des tendances. Ce n’est pas possible avec la qualité et la fréquence des données actuellement disponibles, mais on pourrait peut-être le faire en utilisant des modèles. Nous serions intéressés de connaître toute personne intéressée par la poursuite de ces objectifs, reconnaissant le fait que le GSEAf n’a actuellement pas de ressources pour un tel processus. Ayant numérisé toutes les limites des études où les fournisseurs de données ne nous ont pas donné les informations, Peter a également commencé à travailler sur la nouvelle carte de l’habitat. C’est une partie du processus qui prend beaucoup de temps et nous faisons


Dublin

At the IUCN World Conservation Congress in September, Diane Skinner worked with Simon Hedges, Co-Chair of the Asian Elephant Specialist Group, to convene a Knowledge Café event about the new African and Asian Elephant Database. The objective of the session was to further sensitize other parts of IUCN and other specialist groups on the features, values and potential uses of the system. One useful outcome of the session was a deeper understanding of the process that a specialist group needs to go through to determine which data management tools might be best for them, depending on the nature and characteristics of the population data they are dealing with, as well as the analytical needs they might have. This information has been passed on to the IUCN Secretariat’s species team and, hopefully, will assist in supporting a number of specialist groups to better manage their data in future.

Peter Mwangi’s first year with AfESG also allowed him some new experiences, including participating in two different new events. One was the 2012 survey of the Laikipia–Samburu ecosystem in northern Kenya, where Peter experienced many different parts of the count—and loved it! The second was the workshop IUCN ESARO held on data management needs across East and Southern Africa, part of the larger IUCN project on building capacity for protected area management, also known as BIOPAMA.

Illegal killing and ivory tradeAs I reported in the last Chair report, Diane Skinner and I attended the 62nd meeting of the CITES Standing Committee in Geneva in July. Diane circulated a document explaining the outcomes of that meeting to AfESG members and we have now, like everyone else, turned our attention to the 16th meeting of the Conference of the Parties, to be held next March in Bangkok, Thailand. Many of the processes we have been following and assisting with since CoP15 in Doha will be finalized or progressed at CoP16, including the revision of Resolution Conf. 10.10 (Rev CoP15) and discussions surrounding a possible decision-making mechanism for future trade in ivory.

appel aux fournisseurs de données de nous envoyer, si possible, les limites des habitats en même temps que les rapports de leurs recensements très appréciés.

Lors du Congrès mondial de l’UICN en septembre, Diane Skinner a travaillé avec Simon Hedges, Co-président du Groupe de spécialistes de l’éléphant d’Asie, pour organiser un événement du Café des connaissances sur la nouvelle Base de Données sur les Eléphants d’Afrique et d’Asie. L’objectif de la session était de sensibiliser les autres parties et groupes de spécialistes de l’UICN sur les caractéristiques, les valeurs et les utilisations possibles du système. Un résultat utile de la session était une meilleure compréhension du processus qu’un groupe de spécialistes doit suivre pour déterminer les outils de gestion de données qui pourraient leur convenir le mieux, en fonction de la nature et des caractéristiques des données de population dont ils s’occupent, ainsi que les besoins analytiques qu’ils pourraient avoir. Ces informations ont été transmises à l’équipe du Secrétariat des espèces de l’UICN et nous espérons qu’elles contribueront à aider des groupes de spécialistes à mieux gérer leurs données à l’avenir.

La première année que Peter Mwangi a passée au GSEAf lui a permis d’avoir de nouvelles expériences, y compris la participation à deux expériences différentes. L’une était l’étude de 2012 de l’écosystème de Laikipia-Samburu au nord du Kenya, où Peter a vu les différentes composantes du dénombrement - et il a aimé cela! La seconde était l’atelier du Bureau régional de l’Afrique Orientale et Australe de l’UICN tenu pour étudier les besoins de gestion des données à travers l’Afrique orientale et australe, une partie du grand projet de l’UICN sur le renforcement des capacités pour la gestion des aires protégées, également connu sous le nom de BIOPAMA.

L’abattage illégal et le commerce d’ivoireComme je l’ai signalé dans le rapport précédent du Président, Diane Skinner et moi avons participé à la 62ème réunion du Comité permanent de la CITES à Genève en juillet. Diane a fait circuler un document expliquant les résultats de cette réunion aux membres du GSEAf et comme tout le monde, nous avons maintenant tourné notre attention vers la 16ème réunion de la Conférence des Parties, qui se tiendra en mars à Bangkok en Thaïlande. La plupart des processus que nous avons suivis et aidés depuis la CdP15 de Doha seront finalisés ou avancés à la CdP16, y compris la révision de la résolution Conf. 10.10 (Rev CdP15) et les discussions entourant un possible



Update on the CITES-MIKE and ETIS programmesPhase 2 of CITES-MIKE concluded at the end of 2012 and is now undergoing a final evaluation. The European Commission has approved an interim phase of the programme for 2013 and 2014. AfESG will continue to partner with CITES-MIKE and TRAFFIC, in particular working further on bringing together data and analysis links along the supply chain. In November, I attended a series of meetings in Brussels with the European Commission and the Secretariat of the African, Caribbean and Pacific (ACP) Group of States, along with a number of representatives from African elephant range States, TRAFFIC and the CITES Secretariat to sensitize the ACP to the contributions of AfESG, MIKE and ETIS to elephant conservation, within the context of the African Elephant Action Plan. This is in the hope of lobbying for further support for the new phase the European Commission is supporting for the combined efforts of MIKE, ETIS and AfESG from 2015 onwards.

PachydermPachyderm 51 was our first with Dali Mwagore back at the editing helm. While funding continues to be tight, we anticipate being able to produce at least issues 52 and 53 with the current arrangements in place. We are also receiving a good number of elephant contributions; our major challenge at the moment being to find enough reviewers with the time to help our senior editor, Debbie Gibson, to review the many manuscripts in the pipeline. If you are interested in being a reviewer, please register on http://pachydermjournal.org and also let Debbie and Dali know that you are ready and willing to help out. Your efforts really do contribute to keeping Pachyderm in high demand.

Human–elephant conflictSince the publication of Richard Hoare’s review paper in Pachyderm 51, we have been working on a plan for AfESG’s 2013 work programme with regard to human–elephant conflict. Funding permitting, we plan to hold a meeting that will

mécanisme de prise de décision pour le futur commerce de l’ivoire.

Mise à jour sur les programmes de la CITES-MIKE et ETISLa phase II de CITES-MIKE a été conclue à la fin de 2012 et fait actuellement l’objet d’une évaluation finale. La Commission Européenne a approuvé une phase intermédiaire du programme pour 2013 et 2014. Le GSEAf continuera à collaborer avec CITES-MIKE et TRAFFIC, surtout en travaillant davantage à rassembler les liens des données et des analyses le long de la chaîne d’approvisionnement. En novembre, j’ai participé à une série de réunions à Bruxelles avec la Commission Européenne et le Secrétariat du Groupe des Etats d’Afrique, des Caraïbes et du Pacifique (ACP), avec un certain nombre de représentants des Etats de l’aire de distribution des éléphants d’Afrique, TRAFFIC et le Secrétariat de la CITES pour sensibiliser les pays ACP à la contribution au GSEAf, MIKE et ETIS pour la conservation des éléphants, dans le cadre du Plan d’action pour l’éléphant d’Afrique. Cela est dans l’espoir de faire du lobbying en faveur d’un soutien supplémentaire pour la nouvelle phase que la Commission Européenne soutient pour les efforts combinés de MIKE, ETIS et GSEAf à partir de 2015.

PachydermLe numéro 51 de Pachyderm était le premier avec le retour de Dali Mwagore à la tête de la publication. Bien que le financement continue à être limité, nous prévoyons d’être en mesure de produire au moins les numéros 52 et 53 avec les dispositions actuellement en place. Nous recevons aussi un bon nombre de contributions sur les éléphants; à l’heure actuelle notre défi majeur est de trouver assez de critiques ayant suffisamment de temps pour aider notre Rédactrice de section, Debbie Gibson, à revoir les nombreux manuscrits qui attendent. Si vous êtes intéressé à être un critique, veuillez vous inscrire sur http://pachydermjournal.org et faites également savoir à Debbie et Dali que vous êtes prêt et disposé à aider. Vos efforts contribuent vraiment au maintien de Pachyderm en très forte demande.


Dublin

include the current HEC Working Group, new members who have the necessary expertise, as well as institutional stakeholders such as WWF and FAO, to update the AfESG tools and guidelines, and to determine the future scope and modus operandi of the working group.

ConclusionI am close to finalizing AfESG member appointments for 2013 to 2016. Building on our redesigned website, I look forward to engaging with a re-energized membership. In my last Chair report, I implored AfESG members and the elephant conservation community to continue their efforts on behalf of elephants, and there has certainly been a great deal of excellent work done in the past few months. I thank you all for that. However, in the run-up to CITES CoP16, which has the potential to be highly divisive, I hope that we can all remember that much more unites our elephant conservation community than divides it. I truly hope that we can come together to take urgent and effective action on behalf of Africa’s elephants now and in the future—let’s maintain our passion, our commitment and our camaraderie.

Conflits homme–éléphantDepuis la publication du document de synthèse de Richard Hoare dans Pachyderme 51, nous avons travaillé sur un plan pour le programme de travail du GSEAf pour 2013 en ce qui concerne les conflits homme-éléphant. Si les fonds le permettent, nous avons l’intention d’organiser une réunion qui inclura le Groupe de travail actuel sur le CHE, les nouveaux membres qui possèdent l’expertise nécessaire ainsi que les acteurs institutionnels tels que le WWF et la FAO, pour mettre à jour les outils et les lignes directrices du GSEAf, et déterminer la portée future et le modus operandi du Groupe de travail.

ConclusionJe suis sur le point de finaliser les nominations des membres du GSEAf pour 2013 à 2016. En m’appuyant sur notre nouveau site, je me réjouis de coopérer avec des membres redynamisés. Dans mon dernier rapport du Président, j’ai imploré les membres du GSEAf et la communauté de la conservation des éléphants de poursuivre leurs efforts en faveur des éléphants, et il y a certainement eu beaucoup d’excellent travail accompli au cours de ces derniers mois. Je vous remercie tous pour cela. Cependant, dans la perspective de la CdP16 de la CITES, qui a le potentiel d’être très délicat, j’espère que nous pouvons tous nous rappeler qu’il y a beaucoup plus de choses qui unissent la communauté de la conservation des éléphants que celles qui la divisent. J’espère sincèrement que nous pouvons nous unir pour prendre des mesures urgentes et efficaces au nom des éléphants d’Afrique, maintenant et dans l’avenir – maintenons notre passion, notre engagement et notre camaraderie.


African Rhino Specialist Group report

IUCN World Conservation Congress and rhino motionThe IUCN World Conservation Congress held in Jeju, Republic of Korea, 6–15 September 2012, approved a motion, ‘Conservation of rhinoceros species in Africa and Asia’, which has since been formalized into IUCN Recommendation 138, whose text is given below:

ACKNOWLEDGING that the world’s five species of rhinoceros are charismatic emblems of conservation;

APPRECIATING that effective conservation measures and significant political will and conservation expenditure in some range States in recent years have led to population increases in three species—the Black, Southern White and Greater One-horned Rhinos;

RECOGNIZING the important role that commercial wildlife enterprises, including trophy hunting, have played in generating incentives for conservation and stimulating population increases of rhinos on state, private and communal land in Africa;

ALARMED that the populations of the two rarest species, the Javan and Sumatran rhinos, continue to decline, and are now at perilously low levels;

DISTRESSED that two rhino subspecies, the Western Black Rhino (Diceros bicornis longipes) in Cameroon and the Indochinese Javan rhino (Rhinoceros sondaicus annamiticus) in Viet Nam have gone extinct in the last decade;

AWARE that the Northern White Rhino (Ceratotherium simum cottoni) and the mainland populations of the Sumatran Rhino are now extremely close to extinction;

DEEPLY CONCERNED that pressure from illegal hunting on all species of rhinos has grown

African Rhino Specialist Group reportRapport du Groupe des Spécialistes des Rhinocéros d’Afrique

Mike Knight, Chair/Président

Park Planning and Development, South African National Parks, PO Box 76693, and Centre for African Conservation Ecology, Nelson Mandela Metropolitan University, Port Elizabeth 6013, South Africaemail: [email protected]

Le Congrès Mondial de l’UICN sur la Conservation et la motion sur le rhinocérosLe Congrès mondial de l’UICN de Jeju en République de Corée, du 6 au 15 septembre 2012 a approuvé une motion intitulé Conservation des espèces de rhinocéros en Afrique et en Asie, ce qui a été depuis officialisée dans la recommandation de l’UICN 138, dont le texte figure ci-dessous.

« RECONNAISSANT que les cinq espèces de rhinocéros dans le monde sont des symboles charismatiques de conservation;

CONSCIENTS que des mesures de conservation efficaces et une volonté politique significative et les dépenses de conservation dans certains Etats de l’aire de distribution ont conduit ces dernières années à une augmentation de la population des trois espèces - le rhinocéros noir, le rhinocéros blanc du sud et le grand rhinocéros unicorne;

RECONNAISSANT le rôle important que les entreprises commerciales de la faune, y compris la chasse aux trophées, ont joué dans la création des motivations pour conserver et promouvoir la croissance des populations de rhinocéros sur les terres publiques, privées et communales en Afrique;

ALARME par le fait que les populations des deux espèces les plus rares, les rhinocéros de Java et de Sumatra, continuent à baisser, et sont maintenant à des niveaux dangereusement bas;

DéPLORANT le fait que deux sous-espèces de rhinocéros, le rhinocéros noir de l’ouest (Diceros bicornis longipes) au Cameroun et le rhinocéros indochinois de Java (Rhinoceros sondaicus annamiticus) au Viet Nam ont disparu au cours de la dernière décennie;

SACHANT que le rhinocéros blanc du nord (Ceratotherium simum cottoni) et les populations continentales de rhinocéros de Sumatra sont maintenant très proches de l’extinction;


Knight

seriously in recent years, linked to a significant increase in non-traditional use of rhino horn and a significant rise in the price of rhino horn in Asian markets, especially in Viet Nam and China, as well as a reduction in the capacity and efficiency of some range State conservation authorities to protect their rhinos;

ALARMED that a continued increase in illegal hunting of rhinos and in rhino horn demand could rapidly jeopardize the improvements that have been achieved in the status of Black, White and Greater One-horned Rhinos over the last two decades, and together with inadequate biological management could easily cause the extinction of the Javan and Sumatran Rhinos in the foreseeable future;

NOTING that the measures taken by the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) to bring the illegal trade in rhino products under control still require commitment from key rhino range States and rhino horn consuming countries;

AWARE that for many rhino range States the cost of securing their rhino populations requires significant assistance from both internal and external sources, and that this has increased opportunity costs to general conservation; and

CONCERNED that the increased risks and costs associated with securing rhinos will possibly provide a disincentive for private owners and custodians of rhino in eastern and southern Africa from investing in rhinos and conservation, especially in the major range State South Africa and also in Zimbabwe (where recent allocations of hunting concessions linked to land reform could pose additional economic threats to private conservancies);

1. COMMENDS those rhino range States in Africa and Asia that have taken strong measures to conserve their remaining rhinos and actions against the incentives to kill them illegally for their horns, thus consequently have increasing populations;

2. CALLS ON all range States to give priority to securing their rhino populations, bringing illegal hunting and trade under control, and ensuring that effective deterrents are in place and enforced in order to minimize the levels of illegal hunting and trade, whilst at the same time seeking to create an enabling environment to encourage the continued expansion of the rhino range and rapid growth in rhino numbers;

PROFONDéMENT PRéOCCUPé que la pression de la chasse illégale sur toutes les espèces de rhinocéros a sérieusement augmenté ces dernières années, et qu’elle est liée à une augmentation significative de l’usage non traditionnel de la corne de rhinocéros et d’une hausse importante du prix de la corne de rhinocéros sur les marchés asiatiques, en particulier au Viet Nam et en Chine, ainsi que la réduction de la capacité et de l’efficacité des autorités de conservation de certains états de aire de répartition de protéger leurs rhinocéros;

ALARME par le fait qu’une augmentation continue de la chasse illégale de rhinocéros et la demande pour leurs cornes pourraient rapidement compromettre les améliorations qui ont été obtenues dans la situation du rhinocéros noir, du rhinocéros blanc et du grand rhinocéros unicorne au cours des deux dernières décennies, conjugué avec une gestion biologique inadéquate, pourraient facilement causer l’extinction des rhinocéros de Java et de Sumatra dans un avenir prévisible;

NOTANT que les mesures prises par la Convention sur le Commerce International des espèces de faune et de flore sauvages menacées d’extinction (CITES) pour maitriser le commerce illicite en produits issus des rhinocéros nécessitent encore de l’engagement des principaux Etats de l’aire de distribution de rhinocéros et les pays consommateurs de cornes de rhinocéros;

SACHANT que le coût de la sauvegarde des populations de rhinocéros de nombreux Etats de l’aire de répartition nécessite une aide importante à partir des sources internes et externes, et que, par conséquent, les coûts d’opportunité de la conservation en général ont augmenté, et

CRAIGNANT que les risques accrus et les coûts associés à la sauvegarde des rhinocéros pourraient peut-être décourager les propriétaires privés et les conservateurs de rhinocéros en Afrique orientale et australe d’investir dans la conservation des rhinocéros, surtout dans les majeurs aires de distribution en Afrique du Sud et aussi au Zimbabwe (où de récentes attributions de concessions de chasse liées à la réforme agraire pourraient poser d’autres menaces économiques pour les zones de conservation privées);

1. FELICITE les Etats de l’aire de répartition du rhinocéros d’Afrique et d’Asie qui ont pris des mesures énergiques pour conserver leurs rhinocéros restants et des actions contre les incitations à les tuer illégalement pour leurs cornes, et qui par la suite ont des populations croissantes;

2. APPELLE tous les Etats de l’aire de distribution à donner la priorité à la sauvegarde de leurs populations de rhinocéros, en maitrisant la chasse et le commerce illégaux,



3. ENCOURAGES all rhino range States to manage their rhino populations to achieve rapid growth, with long-term genetic and demographic viability;

4. FURTHER ENCOURAGES range States to evaluate the pros and cons of alternative strategies to determine how best to reduce the illegal trade, black market prices and illegal demand for rhino horn and hence ultimately reduce poaching;

5. APPLAUDS the initiative of the President of Indonesia for proposing the International Year of the Rhino starting June 2012 and supports his government’s emergency actions to save the Javan and Sumatran Rhinos from extinction, that include: establishing a high-level task force of national and international experts on rhino population and habitat management; identifying the most suitable areas for establishing free-ranging rhinoceros populations; allocating sufficient resources to enforce their protection, to maximize the breeding potential of the remaining animals, and to have regular, frequent and intensive monitoring of all rhino populations;

6. ENCOURAGES the government of Malaysia to take urgent actions to save the Sumatran Rhinoceros population in Sabah from extinction, including through close management of rhinos in fenced, managed conditions, and exploring all possible techniques that may boost birth rate above natural death rate, including super-ovulation, artificial insemination, in vitro fertilization and other advanced reproductive techniques;

7. COMMENDS the governments of India and Nepal for the measures they have taken to secure the status of the Greater One-horned Rhinoceros in its wild habitats, but urges them to establish new, viable, strictly protected populations of the species in previously occupied habitats, as well as enhancing the protection of existing populations, noting that in the case of India this will require concerted action from the Union Government as well as from the State governments of Assam, West Bengal, Bihar and Uttar Pradesh;

8. URGES all Asian rhino range countries to adopt robust scientific techniques to estimate their rhino populations, and to repeat these censuses at least once every two years, ensuring independent peer review of the methods and results;

9. CALLS ON African range States to:

et en veillant à ce que des mesures dissuasives efficaces soient en place et appliquées afin de minimiser les niveaux de chasse et de commerce illégaux, et en même temps en cherchant à créer un environnement favorable pour encourager la poursuite de l’expansion de l’habitat du rhinocéros et une croissance rapide du nombre de rhinocéros;

3. ENCOURAGE tous les Etats de l’aire de distribution de rhinocéros à gérer leurs populations de rhinocéros pour atteindre une croissance rapide ainsi qu’une viabilité génétique et démographique à long terme;

4. ENCOURAGE EN OUTRE les Etats de l’aire de répartition à évaluer les avantages et les inconvénients des différentes stratégies pour déterminer la meilleure façon de réduire le commerce illégal, les prix du marché noir et la demande illicite de cornes de rhinocéros et ainsi à la longue réduire le braconnage;

5. SALUE l’initiative du Président de l’Indonésie de proposer l’Année internationale du Rhinocéros à partir de juin 2012 et de soutenir les actions d’urgence de son gouvernement pour sauver les rhinocéros de Java et de Sumatra d’extinction, actions qui comprennent: la création d’un groupe de travail de haut niveau d’experts nationaux et internationaux sur la population de rhinocéros et la gestion de l’habitat, l’identification des zones les plus appropriées pour l’établissement de populations de rhinocéros en liberté; l’allocation des ressources suffisantes pour faire respecter leur protection, afin de maximiser le potentiel de reproduction des animaux restants, et un suivi régulier, fréquent et intensif de toutes les populations de rhinocéros;

6. ENCOURAGE le Gouvernement de la Malaisie à prendre des mesures urgentes pour sauver la population des rhinocéros de Sumatra en voie d’extinction à Sabah, par exemple, grâce à une gestion stricte des rhinocéros dans les conditions clôturées et gérées, et l’exploration de toutes les techniques possibles qui peuvent stimuler un taux de natalité supérieur au taux de décès naturel, comprenant la super-ovulation, l’insémination artificielle, la fécondation in vitro et d’autres techniques avancées de reproduction;

7. FELICITE les gouvernements de l’Inde et du Népal pour les mesures qu’ils ont prises pour préserver la situation des grands rhinocéros unicornes dans leurs habitats naturels, mais les exhorte à établir de nouvelles populations, viables et strictement protégées dans les habitats occupés précédemment, ainsi qu’à améliorer la protection des populations existantes, en notant que dans le cas de l’Inde, cela nécessitera une action concertée du gouvernement de l’Union ainsi que les gouvernements des Etats de l’Assam, du Bengale occidental, du Bihar et de l’Uttar Pradesh;


Knight

a. increase collaborative law enforcement actions between range States, transit and consuming countries;

b. improve detection of rhino horn at ports of entry/exit with, inter alia, the aid of sniffer dogs, specialized equipment and resourced staff;

c. increase the allocation of national resources towards improving rhino security and conservation authorities’ skills base;

d. increase the focus on intelligence gathering and analysis to stop poachers before killing rhinos;

e. increase the rate of successful prosecutions with deterrent sentences for illegal rhino-related activities;

f. maintain enabling land-use and investment policies together with support for appropriate and well-managed, sustainable, income-generating options that encourage investment in rhinos, sustainable populations and which help fund effective conservation by the private wildlife industry and communities;

g. enhance socio-economic stability through increased local community involvement;

h. encourage private rhino owners to willingly cooperate with the authorities in the provision of rhino information;

i. encourage all range States, transit and consuming countries to improve and regularly supply rhino-related information to the African Rhino Specialist Group (AfRSG) of the IUCN Species Survival Commission (SSC) and to TRAFFIC to facilitate data management, information sharing and their mandated reporting to CITES’ Conference of the Parties;

j. further expand the use of DNA profiling of rhino horns (using the RhoDIS database in Africa) as an innovative means of combating the illegal killing of rhinos and the trafficking of horn;

k. encourage the involvement of civil society in facilitating awareness and generating resources in line with priority needs, as well as assisting with population monitoring, and with implementing rhino conservation; and

l. improve rhino population monitoring to inform management for population growth;

10. RECOGNIZES that the successful conservation of rhinos across the entire range will be best achieved via a diversity of management and economic mechanisms;

8. EXHORTE tous les pays de l’aire de distribution du rhinocéros d’Asie à adopter des techniques scientifiques robustes pour estimer leurs populations des rhinocéros, à répéter ces recensements au moins une fois tous les deux ans, et à assurer un examen des méthodes et des résultats par les pairs indépendants;

9. INVITE les Etats de l’aire de répartition en Afrique à:a. accroître les actions collaboratives de mise en

application de la loi entre les Etats de l’aire de distribution, les pays de transit et les pays consommateurs;

b. améliorer la détection de la corne de rhinocéros dans les ports d’entrée/sortie, notamment avec l’aide de chiens renifleurs, du matériel spécialisé et un personnel ayant des ressources;

c. augmenter l’allocation des ressources nationales vers l’amélioration de la sécurité des rhinocéros et les compétences des autorités de conservation;

d. mettre l’accent sur la collecte et l’analyse de renseignements pour arrêter les braconniers avant qu’ils tuent les rhinocéros;

e. augmenter le taux de réussite des poursuites judiciaires avec des peines dissuasives contre les activités illégales liées aux rhinocéros;

f. maintenir des politiques d’utilisation des terres et d’investissement favorables et soutenir des options appropriées génératrices de revenus, bien gérées et durables qui encouragent l’investissement dans les rhinocéros, des populations viables et qui aident à appuyer financièrement une conservation efficace par le secteur privé de la faune et les communautés ;

g. renforcer la stabilité socio-économique grâce à une participation accrue de la communauté locale;

h. encourager les propriétaires privés de rhinocéros à coopérer avec les autorités dans la provision d’informations sur les rhinocéros;

i. encourager tous les Etats de l’aire de distribution, les pays de transit et les pays consommateurs à améliorer et à fournir régulièrement des informations relatives au rhinocéros, au Groupe de Spécialistes du Rhinocéros d’Afrique de la Commission de la Survie des Espèces de l’UICN et à TRAFFIC pour faciliter la gestion des données, l’échange des informations et leur signalement obligatoire à la Conférence des Parties de la CITES;

j. étendre davantage l’utilisation de profils d’ADN des cornes de rhinocéros (en utilisant la base de données Rhodis en Afrique) comme un moyen innovateur de lutte contre l’abattage illégal de rhinocéros et le trafic de la corne;

k. encourager la participation de la société civile pour faciliter la sensibilisation et la création de ressources en



11. REQUESTS the Director General and SSC (especially its African and Asian Rhino Specialist Groups) to provide ongoing support and guidance to all rhino range States, and in particular to assist in raising the profile of rhinoceros conservation during the International Year of the Rhino;

12. CALLS ON donors to make the necessary financial resources available to enable the range States to secure their rhino populations, and

13. CALLS UPON those States implicated in the increase in demand for rhino horn and the surge in the black market prices for horn to fully cooperate at all levels with the rhino range States in positively addressing the rhino crisis and seeking lasting solutions.

Importantly, this resolution highlights the successes in rhino conservation and the reasons for it, but it also emphasizes the current poaching plight facing rhinos, driven by increasing demand for illegal rhino horn, principally in Viet Nam and China, and the way forward. It also emphasizes the current risks of private rhino owners disinvesting in rhino conservation, which would translate into a further reduction in numbers and homes for rhinos, given that most state reserves in southern Africa are at capacity. The rhino crisis was further highlighted at an IUCN SSC Species Pavilion lunchtime event organized by IUCN’s East and Southern African Regional Office, The AfRSG scientific officer, Richard Emslie, and AfESG’s programme officer, Diane Skinner, gave presentations on the status and challenges facing rhino and elephant conservation in Africa, followed by a discussion session with the audience.

Joint IUCN/TRAFFIC report for CoP16The third joint IUCN SSC/AfRSG/AsRSG and TRAFFIC rhino report has been completed and was forwarded to the CITES Secretariat. It has been appended as an annex to the report on rhinos by the CITES Secretariat and can be downloaded from http://www.cites.org/eng/cop/16/doc/E-CoP16-54-02.pdf. As with previous reports, it is information rich and concludes with a series of recommendations that Parties may wish to consider putting into Decisions at the CoP. AfRSG also intends to prepare a follow-up information

fonction des besoins prioritaires, et aider à la surveillance de la population et à mettre en œuvre la conservation des rhinocéros, et

l. améliorer la surveillance de la population des rhinocéros afin d’informer la gestion de la croissance de leur population;

10. RECONNAIT que le succès de la conservation des rhinocéros dans tout l’habitat se réalisera le mieux par une diversité de mécanismes de gestion et économiques;

11. PRIE le Directeur général et la Commission de la Survie des Espèces (surtout ses Groupes de Spécialistes du Rhinocéros d’Afrique et d’Asie) de fournir un soutien continu et des conseils à tous les Etats de l’aire de distribution des rhinocéros, et en particulier pour aider à rehausser le profil de la conservation du rhinocéros au cours de l’Année internationale du Rhinocéros;

12. INVITE les bailleurs de fonds à assurer les ressources financières nécessaires pour permettre aux Etats de l’aire de distribution de sécuriser leurs populations de rhinocéros restants, et

13. INVITE les Etats impliqués dans l’augmentation de la demande pour la corne de rhinocéros et la flambée des prix du marché noir de la corne à coopérer pleinement à tous les niveaux avec les Etats de l’aire de répartition des rhinocéros pour faire face positivement à la crise du rhinocéros et à la recherche de solutions durables. »

Fait important, cette résolution met en exergue les réussites en matière de conservation des rhinocéros, mais insiste aussi sur le problème actuel du braconnage auquel les rhinocéros sont confrontés, causé par la demande croissante de cornes illégales de rhinocéros principalement au Viet Nam et en Chine, et la voie à suivre. Elle souligne également le risque actuel que les propriétaires privés des rhinocéros pourraient se désinvestir de la conservation des rhinocéros ce qui se traduirait par encore une réduction des rhinocéros et de leurs habitats, étant donné que la plupart des réserves d’Etat de l’Afrique australe sont à pleine capacité. La crise du rhinocéros a en outre été mise en exergue lors d’un déjeuner au Pavillon de la CSE de l’UICN organisé par le Bureau Régional de l’Afrique Orientale et Australe de l’UICN. Richard Emslie, spécialiste des questions scientifiques au GSRAf et Diane Skinner, chargée de programme au GSEAf ont fait des exposés sur la situation et les défis auxquels la conservation des rhinocéros et des éléphants est confrontée en Afrique, suivis d’une séance de discussion avec le public.


Knight

document at CITES CoP16 to provide Parties with the latest information on rhino numbers, trends and trade following its February 2012 meeting.

Poaching updateIn the major rhino range State, South Africa, poaching has markedly escalated during the reporting period, reaching new highs in the last three months of the year with a rhino being poached almost every nine hours (Fig. 1).

The decision to prevent nationals of Viet Nam from obtaining hunting licences, and changes to the law in April 2012 requiring a law-enforcement official to attend every hunt and prospective hunters to demonstrate that they are bona fide, genuine sport hunters, have significantly constricted the illicit rhino horn supply from pseudo-hunting. This constriction in supply could have forced organized criminals to look for alternative sources of illegal horn to supply unlawful demand, such as from more poaching (by far the largest source of illegal horn), illegal dehorning, or theft. However, other factors that are completely unrelated to South Africa’s policy decisions and legislative and law-enforcement changes such as increasing corruption, the emergence of new markets or the escalation in existing demand, could also be behind these increases.

While current poaching levels in South Africa

Rapport conjoint de l’UICN/TRAFFIC pour la CdP16Le troisième rapport conjoint de la CSE de l’UICN/le GSRAf/le GSRAs et TRAFFIC sur le rhinocéros a été finalisé et transmis au Secrétariat de la CITES. Il a été joint en annexe au rapport sur les rhinocéros du Secrétariat de la CITES et on peut le télécharger à partir de http://www.cites.org/eng/cop/16/doc/E-CoP16-54-02.pdf . Comme les rapports précédents, il est riche en informations et se termine par une série de recommandations que les Parties pourraient envisager d’inclure dans les décisions de la CdP. Le GSRAf a aussi l’intention de préparer un document d’informations de suivi pour la CdP16 de la CITES afin de fournir aux Parties les dernières informations sur le nombre de rhinocéros, les tendances et le commerce suite à sa réunion de février 2012.

Mise à jour sur le braconnageDans le majeur Etat de l’aire de répartition du rhinocéros, l’Afrique du Sud, le braconnage s’est considérablement intensifié au cours de la période considérée pour atteindre de nouveaux records au cours des trois derniers mois de l’année avec un rhinocéros braconnés presque toutes les neuf heures (Fig. 1).

La décision d’empêcher les ressortissants du Viet Nam d’obtenir un permis de chasse, et les modifications apportées à la loi en avril 2012 exigeant qu’un agent de la loi assiste à chaque chasse, et exigeant que les chasseurs potentiels

Figure 1. The number of rhinos poached per day in South Africa, by year and quarter. [Le nombre de rhinocéros braconnés par jour en Afrique du Sud, par année et par trimestre.]

0.040.23

0.33

0.91

1.23

1.83

0.56

0.901.01

1.17 1.17 1.16 1.14

1.481.69

1.15

1.86

2.61

0.00

0.50

1.00

1.50

2.00

2.50

3.00

Avg/

yr 1

8 yr

s 1

990-

2007 20

08

2009

2010

2011

Jan-

19 D

ec 2

012

Jan-

Mar

201

0Ap

r-Jun

201

0Ju

l-Sep

t 201

0O

ct-D

ec 2

010

Jan-

Mar

201

1Ap

r-Jun

201

1Ju

l-Sep

t 201

1O

ct-D

ec 2

011

Jan-

Mar

201

2Ap

r-Jun

e 20

12Ju

ly-S

ep t2

012

Oct

-Dec

201

2

Average annual ratesTaux annuels moyens

2010 qtr 2012 qtr 2011 qtr

No.

of r

hino

s po

ache

d pe

r day

Nom

bre

de rh

inoc

éros

bra

conn

ées

par j

our

0.00

0.50

1.00

1.50

2.00

2.50

3.00

Period



are still well below historical metapopulation growth rates, should the escalation in poaching continue deaths will eventually start to exceed births, with rhino numbers starting to decline in South Africa as early as 2015/16. The escalation in poaching has also significantly increased the costs of protecting rhinos as well as increasing risks to rhinos and their owners, families and field staff. Live sale prices have been declining further, reducing incentives to the private rhino industry. This has coincided with an increasing number of white rhino owners getting rid of or thinking about getting rid of their rhinos. If this trend escalates it is of major concern given that the ultimate carrying capacity for rhinos and new homes for surplus white rhinos are dependent upon the private sector and community willingness to conserve them. Any decline in live sales also threatens to reduce budgets for conservation as well as negatively affect biological management.

In the first nine months of the year poaching encouragingly was down in Kenya and Zimbabwe, with some animals lost in Malawi and Tanzania. However, in the last two months of 2012, there has been a sharp escalation in poaching in Kenya with a total of 15 rhinos lost, bringing the annual total to 29 poached rhinos. With a total population of 1,000+ rhinos in Kenya, this poaching offtake translated to about 3% of the entire population. Overall, Kenya lost 103 (71 black and 32 white rhinos) in a five-year period between 2008 and 2012 (Fig. 2). Not only is the impact considered to be severe on the population growth trend, it is also negatively affecting the hectarage under rhino conservation. The joint IUCN/TRAFFIC report for CoP16 contains more information including a table documenting rhino poaching by country by year since 2006.

More convictions with deterrent sentencesIn my last report I mentioned that the number of convictions with deterrent sentences was increasing. This trend has continued in South Africa. A man convicted of illegally dehorning and trying to illegally sell the horns was sentenced to eight years in jail, plus a 1 million rand fine (~$117,235) to be paid to the Environmental Management Inspectorate (‘Green Scorpions’).

prouvent qu’ils sont de bonne foi, c’est-à-dire qu’ils sont de véritables chasseurs sportifs, ont considérablement réduit l’offre illicite de cornes de rhinocéros provenant de la pseudo-chasse. Cette réduction de l’offre aurait forcé les criminels organisés à chercher d’autres sources de cornes illégales pour répondre à la demande illégale, telles que celles provenant du braconnage accru (de loin la plus importante source des cornes illégales), de l’écornage illégal ou des vols. Cependant, d’autres facteurs qui n’ont aucun rapport avec les décisions politiques en Afrique du Sud et les changements législatifs et d’application de la loi tels que la corruption croissante, l’émergence de nouveaux marchés ou l’intensification de la demande existante, pourraient également être à l’origine de ces augmentations.

Bien que les niveaux actuels de braconnage en Afrique du Sud soient encore en deçà des taux de croissance historiques des métapopulations, si l’intensification du braconnage devait continuer, éventuellement les décès commenceraient à dépasser les naissances et le nombre de rhinocéros à diminuer en Afrique du Sud dès 2015/16. L’intensification du braconnage a également augmenté de façon significative les coûts de protection des rhinocéros ainsi que les risques encourus par les rhinocéros et leurs propriétaires, les familles et le personnel de terrain. Le prix des animaux vivants a encore diminué, réduisant les incitations à l’industrie privée de rhinocéros. Cela a coïncidé avec le moment où un nombre croissant de propriétaires se débarrassent ou pensent se débarrasser de leurs rhinocéros blancs. Si cette tendance s’intensifie, c’est très préoccupant étant donné que la capacité de charge pour les rhinocéros et les nouveaux habitats pour les rhinocéros blancs excédentaires dépendent du secteur privé et de la volonté de la communauté de les conserver. Toute baisse des ventes des animaux vivants menace aussi de réduire les budgets de conservation et d’avoir des répercussions négatives sur la gestion biologique.

Au cours des neuf premiers mois de l’année, le braconnage a heureusement diminué au Kenya et au Zimbabwe, avec la perte de quelques animaux au Malawi et en Tanzanie. Toutefois, dans les deux derniers mois de 2012, il y a eu une brusque intensification du braconnage au Kenya avec la perte d’un total de 15 rhinocéros, ce qui porte le total annuel à 29 rhinocéros braconnés. Avec une population totale de plus de 1.000 rhinocéros au Kenya, ce prélèvement dû au braconnage représente près de 3% de la population totale. Dans l’ensemble, le Kenya a perdu 103 rhinocéros (71 rhinocéros noirs et 32 rhinocéros blancs) sur une période de cinq ans entre 2008 et 2012 (Fig. 2). Non seulement l’impact est considéré comme grave pour


Knight

He also forfeited additional assets to the value of over $1.3 m (over 11m rand). The Thai criminal Chumlong Lemtongthai, who used Thai sex workers in Gauteng to apply for white rhino hunting permits and was actively using psuedo-hunting as a vehicle to obtain horns for illegal markets, was sentenced to 40 years in jail; although he has appealed. Kenya’s cabinet ministers recommended a draft wildlife conservation and management bill in October 2012 for parliamentary debate. The draft bill contains proposed penalties that are more severe than the current penalties for wildlife-related crimes. The bill was due for debate by the current Kenya’s parliament, whose term ended on 14 January 2013.

TRAFFIC Viet Nam reportA major TRAFFIC report on the rhino horn trade between South Africa and Viet Nam written by Tom Milliken and Jo Shaw was released during the reporting period. It highlights Viet Nam’s increasing role in illegal rhino horn trade as well as South Africa’s responses. It makes for recommended reading and can be downloaded from http://www.rhinoresourcecenter.com/pdf_files/134/1345588780.pdf. A shorter summary document is also available at http://www.rhinoresourcecenter.com/pdf_files/134/1345590401.pdf. The report concludes with a number of recommendations. Encouragingly, these recommendations were subsequently discussed together with progress in addressing them by South Africa’s Department of Environmental Affairs

la croissance de la population, mais cela affecte aussi de façon négative la superficie pour la conservation des rhinocéros. Le rapport conjoint UICN/TRAFFIC pour la CdP16 contient de plus amples renseignements, y compris un tableau indiquant le braconnage de rhinocéros par pays et par année depuis 2006.

Plus de condamnations à des peines dissuasivesDans mon dernier rapport, j’ai mentionné que le nombre de condamnations à des peines dissuasives augmentait. Cette tendance s’est poursuivie en Afrique du Sud. Un homme reconnu coupable d’avoir illégalement écorné et essayé de vendre illégalement des cornes a été condamné à huit ans de prison, plus une amende de 1 million de rands (~117 235$) à verser à l’Inspectorat de la Gestion de l’Environnement (« Scorpions Verts »). On a également confisqué des biens additionnels d’une valeur de plus de $1,3 millions (plus de 11 millions de rands). Un Thaï, Chumlong Lemtongthai, qui utilisait les prostituées thaïlandaises dans le Gauteng pour faire la demande de permis de chasse de rhinocéros blancs et utilisait activement la pseudo-chasse comme un moyen d’obtenir des cornes pour les marchés illégaux a été condamné à 40 ans de prison; mais il a fait appel. Le Conseil des Ministres du Kenya a recommandé un projet de loi à soumettre au Parlement sur la conservation et la gestion de la faune en octobre 2012. Le projet de loi prévoit des sanctions proposées qui sont plus sévères que les peines actuelles relatives aux crimes de la faune. Le Parlement actuel du Kenya, dont le mandat a expiré le 14 janvier 2013, devait débattre ce projet de loi.

Rapport de TRAFIC sur le Viet NamUn rapport important de TRAFFIC sur le commerce de cornes de rhinocéros entre l’Afrique du Sud et le Viet Nam écrit par Tom Milliken et Jo Shaw a été publié au cours de la période considérée. Il souligne le rôle croissant du Viet Nam dans le commerce illégal des cornes de rhinocéros ainsi que les réponses de l’Afrique du Sud. On recommande sa lecture et on peut le télécharger à partir de http://www.rhinoresourcecenter.com/pdf_files/134/1345588780.pdf. Un document plus court est également disponible sur http://www.rhinoresourcecenter.com/pdf_files/134/1345590401.pdf. Le rapport se termine par un certain nombre de recommandations. Fait encourageant, ces recommandations ont ensuite été discutées ainsi que les progrès réalisés dans leur mise en œuvre par le Ministère des affaires

Figure 2. Rhino poaching trend in Kenya from 2007 to 2012. [Tendance du braconnage des rhinocéros au Kenya de 2007 à 2012.] Source: Kenya Wildlife Service / Service Kenyan de la Faune.

02007

10

white rhino (rhino blanc)black rhino (rhino noir)

5

1

5

16

6

16

11

14

9

20

2008 2009

Year / Année

No.

of r

hino

s po

ache

d /

nom

bre

de rh

inoc

éros

bra

conn

ées

2010 2011 2012

5

10

15

20

25

30

35



in a report to a meeting of the South African Parliamentary Portfolio Committee on Water and Environmental Affairs in November 2012.

South African rhino issues management dialogue process continuesIn my last Pachyderm report, I mentioned that a rhino issues manager (RIM), Mr Mavuso Msimang, had been appointed by South Africa’s Department of Environmental Affairs, and together with his team he was coordinating a consultative rhino issues process with a view to informing debate and ultimately coming up with a set of recommendations for the minister. I reported on the first two public meetings as part of this RIM process in my last report. During this reporting period, there have been another six public meetings in Johannesburg and Umhlanga. One meeting discussed biological management and other aspects of conservation, while there were two separate meetings on security and another two on trade issues. There was also a final public summary meeting. The AfRSG Secretariat and other AfRSG members assisted on request, giving a number of background and summary presentations at the meetings, and providing technical information on request to the RIM team. In addition to these public meetings, Mr Msimang also held a number of additional one-on-one meetings as well as a number of meetings with the diplomatic community. The RIM team reported back on their findings and the likely recommendations to senior staff from the formal conservation agencies but this was not a public meeting. A draft report with recommendations has been submitted to the minister. The minister requested further information and is expected to present the report to parliament in early 2013.

Next AfRSG meetingPlans and organizations are well under way for the 11th meeting of AfRSG, which is being held in Naro Moru, Kenya. I will report on this meeting in the next issue of Pachyderm.

environnementales de l’Afrique du Sud lors d’une réunion d’une commission parlementaire sud-africaine sur l’eau et l’environnement en novembre 2012.

Le processus de dialogue sur la gestion des questions relatives au Rhinocéros d’Afrique du Sud se poursuitDans mon dernier rapport pour Pachyderme, j’ai mentionné qu’un Gestionnaire des Questions relatives au Rhinocéros (GQR), Mr. Mavuso Msimang, avait été nommé par le Ministère sud-africain des affaires environnementales et en collaboration avec son équipe, il a coordonné un processus consultatif sur les questions relatives aux rhinocéros en vue d’éclairer le débat et, finalement, proposer une série de recommandations au Ministre. J’ai rendu compte des deux premières réunions publiques dans le cadre de ce processus du GQR dans mon dernier rapport. Pendant la période considérée, il y a eu encore six réunions publiques à Johannesburg et à Umhlanga. Une réunion a porté sur la gestion biologique et d’autres aspects de la conservation, tandis que deux réunions ont porté sur la sécurité et deux autres sur les questions commerciales. Il y avait aussi une réunion publique de synthèse finale. Sur demande, le Secrétariat du GSRAf et les autres membres du GSRAf ont aidé à donner des présentations sur l’historique lors des réunions, et à fournir des informations techniques à l’équipe du GQR. En plus de ces réunions publiques, Mr. Msimang a également tenu un certain nombre d’autres réunions en tête-à-tête ainsi que des réunions avec la communauté diplomatique. L’équipe du GQR a rendu compte de ses conclusions et ses possibles recommandations aux cadres supérieurs des organismes officiels de conservation, mais ce n’était pas une réunion publique. Un projet du rapport contenant des recommandations a été soumis au Ministre. Le Ministre a demandé de plus amples informations et il prévoit de présenter le rapport au Parlement au début de 2013.

Prochaine réunion du GSRAfLes préparatifs pour la 11ème réunion du GSRAf qui se tiendra à Naro Moru au Kenya vont bon train. Je vais présenter un rapport sur cette réunion dans le prochain numéro de Pachyderm.


Knight

Technologie de la Namibie et réunion sur la sécuritéLe Service de la Pêche et de la Faune des Etats Unis a invité 40 experts de tous les Etats de l’aire de répartition des rhinocéros du 25 au 29 novembre 2012 au Parc du plateau de Waterberg en Namibie pour aborder la sécurité des rhinocéros. La réunion était co-sponsorisée par la générosité du Ministère namibien de l’Environnement et du Tourisme et Save the Rhino International (SRI). Les participants comprenaient des gestionnaires de la faune, des praticiens de terrain, des experts en sécurité et les agents d’application de la loi travaillant sur la conservation des rhinocéros au Kenya, au Malawi, en Namibie, en Afrique du Sud, au Swaziland, en Tanzanie, en Zambie et au Zimbabwe. Les participants ont échangé leurs expériences sur les meilleurs moyens de protéger les populations des rhinocéros d’Afrique et ont identifié d’éventuelles interventions diplomatiques et l’équipement technologique qui pourraient améliorer considérablement la survie des rhinocéros.

La rencontre a été l’occasion de 1) présenter les praticiens sur le terrain à d’autres personnes confrontées à des défis similaires, 2) permettre aux gens sur le terrain de partager les connaissances sur les techniques et les technologies qui marchent et celles qui ne marchent pas et dans quelles conditions/circonstances, et 3) donner des conseils au Service de la Pêche et de la Faune des Etats-Unis et d’autres agences du gouvernement américain sur la meilleure façon de contribuer à l’amélioration de la sécurité du rhinocéros grâce à la technologie ou des politiques. La réunion comprenait également des séances pratiques pour démontrer et tester de nouvelles technologies, y compris les drones et les équipements pour le marquage et la localisation des rhinocéros.

A la suite des discussions lors de la réunion, le Service de la Pêche et de la Faune des Etats-Unis a dressé une liste des possibilités de changements de politiques aux niveaux international, régional et national dans les Etats de l’aire de répartition des rhinocéros, et aussi une liste des besoins en équipement standard pour les aires de distribution de rhinocéros.

Plans nationaux pour les rhinocérosPendant la période considérée, le Kenya a officiellement lancé sa quatrième édition de la stratégie nationale de conservation et de gestion du rhinocéros noir. On peut le télécharger à partir de http://www.kws.org/export/sites/kws/info/publications/strategies/CONSERVATION_

Namibian technology and security meetingThe US Fish and Wildlife Service (USFWS) convened 40 experts from across the African rhino range States from 25 to 29 November 2012 in Waterberg Plateau Park in Namibia to address rhino security. The meeting was generously co-sponsored by the Namibian Ministry of Environment and Tourism and Save the Rhino International (SRI).

Participants included wildlife managers, field practitioners, security experts and law-enforcement officers working on rhino conservation in Kenya, Malawi, Namibia, South Africa, Swaziland, Tanzania, Zambia and Zimbabwe. The meeting shared experiences on the best ways to protect Africa’s rhino populations and identified possible diplomatic interventions and technological equipment that could vastly improve rhino survival.

The meeting provided the opportunity to 1) introduce field practitioners to others facing similar challenges, 2) allow field people to share knowledge on which techniques and technologies are working and which are not, and under what conditions or circumstances, and 3) advise USFWS and other US government agencies on how best to contribute to improved rhino security through technology or policy. The meeting also included practical sessions to demonstrate and test new technologies including unmanned aerial vehicles and rhino marking and tracking equipment.

As a result of discussions at the meeting, USFWS has compiled an outline of opportunities for policy changes at international, regional and national levels in rhino range States and a list of standard equipment needs for rhino areas.

National rhino plansDuring the reporting period, Kenya officially launched its fourth edition of the national conservation and management strategy for the black rhino in Kenya. It can be downloaded from http://www.kws.org/export/sites/kws/info/publications/strategies/CONSERVATION_AND_MANAGEMENT_STRATEGY_FOR_THE_BLACK_RHINO_IN_KENYA_2012-2016.



pdf. The overall goal of this strategic plan is to increase black rhino numbers by at least 5% per annum with poaching reduced significantly and additional areas secured for population expansion to reach a confirmed total of 750 rhinos by the end of 2016. Six strategic objectives—Protection and law enforcement; Monitoring for management; Biological management; Population expansion; Awareness and public support; and Coordination and capacity—are outlined to achieve the overall aim. A new national Rhino Steering Committee was set up to ensure the strategic plan is implemented.

A number of other revised national plans still await official sign off. Hopefully some will have been signed off before the AfRSG meeting in February 2013.

The revised South African black rhino biodiversity management plan has been signed by the minister and awaits gazetting in January 2013. A draft South African national white rhino plan has also been completed following a stakeholder workshop and is now undergoing further editing by the SADC Rhino Management Group.

The revised Botswana plan has yet to be finalized and approved as the Department of Wildlife and National Parks is sorting out its policy towards private sector rhino ownership. However, we hope this plan will be finalized and approved soon.

Although completed some time ago, the revised Zimbabwe black rhino conservation strategy also has not yet been signed by the minister of the Environment.

Rhino and Elephant Security Group and INTERPOL Environmental Crime Working Group meeting

A productive meeting of the Rhino and Elephant Security Group and the Interpol Environmental Crime Working Group was held at the Selous Bush Camp near Pilanesberg National Park in South Africa from 8 to 13 October 2012. In light of the increase in elephant and rhino poaching, it was decided to extend invitations to non-SADC countries Kenya and Uganda, both of which sent representatives. In addition to the usual reports, presentations and discussions covering a range of security issues, the meeting had a useful

AND_MANAGEMENT_STRATEGY_FOR_THE_BLACK_RHINO_IN_KENYA_2012-2016.pdf. L’objectif global de ce plan stratégique est d’augmenter le nombre de rhinocéros noirs d’au moins 5% par an en réduisant le braconnage de manière significative et en sécurisant d’autres zones pour l’expansion de la population pour atteindre un total confirmé de 750 rhinocéros à la fin de 2016. Six objectifs stratégiques - la protection et l’application des lois; la surveillance pour la gestion, la gestion biologique, l’augmentation de la population; la sensibilisation et le soutien du public et la coordination et le renforcement sont décrits pour qu’on puisse atteindre l’objectif global. Un nouveau Comité directeur national sur le Rhino a été créé pour assurer la mise en œuvre du plan stratégique.

Plusieurs autres plans nationaux révisés attendent toujours l’approbation officielle. J’espère que certains auront été signés avant la réunion du GSRAf en février 2013.

La version révisée du plan de gestion de la biodiversité du rhinocéros noir d’Afrique du Sud a été signée par le Ministre et attend la publication au Journal officiel en janvier 2013. Un projet de plan national du rhinocéros blanc d’Afrique du Sud a également été réalisé à la suite d’un atelier des parties prenantes et il fait actuellement l’objet d’autres modifications par le Groupe de gestion des rhinocéros de la SADC.

Le plan révisé du Botswana n’a toujours pas été finalisé et approuvé car le Département de la faune et des parcs nationaux est en train de repenser sa politique à l’égard de la possession des rhinocéros par le secteur privé. Cependant, nous espérons que ce plan sera bientôt finalisé et approuvé.

Bien que la version révisée de la stratégie de conservation des rhinocéros noirs au Zimbabwe soit terminé il y a quelque temps, elle non plus n’a toujours pas été signée par le Ministre de l’Environnement.

Réunion du Groupe sur la Sécurité du Rhinocéros et de l’Eléphant/et le Groupe de Travail d’Interpol sur le Crime de l’Environnement

Une réunion productive du Groupe sur la Sécurité du Rhinocéros et de l’Eléphant/et le Groupe de Travail d’Interpol sur le Crime de l’Environnement a eu lieu au Camp Selous Bush près du Parc National de Pilanesberg en Afrique du Sud du 8 au 13 octobre 2012. Etant donné l’augmentation du braconnage des éléphants et


Knight

workshop on intelligence database options. After a presentation, delegates also had a useful practical session on DNA sample collection using the special forensic kits available.

62nd meeting of the CITES Standing CommitteeThe AfRSG scientific officer attended the 62nd meeting of the CITES Standing Committee and took part in a brief CITES Rhino Working Group document. The scientific officer and Chair also provided an updated report, which IUCN printed and made available to delegates. The AfRSG Secretariat also commented on a draft CITES Rhino Working Group report for CITES CoP16 that was produced just before the submission deadline by the outgoing working group chair. Unfortunately, concerns expressed by AfRSG regarding the declining incentives for private sector owners and communities to conserve rhino and the worrying increase in owners disinvesting in rhinos were not captured in the final version that was submitted to the CITES Secretariat. The report stresses the need to reduce demand. Hopefully there will be opportunities for more in-depth discussion at future meetings of the working group, as well as intersessionally.

Other conferencesDuring the reporting period the AfRSG Chair and scientific officer gave invited presentations and were part of a panel discussing rhino conservation challenges and approaches at the International Wildlife Management Congress in Durban in July 2012. In September the Chair also attended and spoke at the Rhino Technical Advisory Group session at the annual European Association of Zoos and Aquaria conference in September 2012 in Innsbruck, Austria. The precarious situation facing Africa’s rhinos was presented. A call was also made for greater cooperation between zoo and conservation communities, especially towards supporting key field-focused rhino conservation programmes. The scientific officer also gave a keynote address at the Symposium of Contemporary Conservation Practice held in KwaZulu-Natal, South Africa, in October 2012.

des rhinocéros, il a été décidé d’étendre les invitations aux pays qui n’appartiennent pas à la SADC comme le Kenya et l’Ouganda, qui ont tous les deux envoyé des représentants. En plus des rapports habituels de feedback, des présentations et des discussions portant sur un éventail de questions de sécurité, la réunion comprenait un atelier utile sur les options de base de données sur les renseignements. Après une présentation, les délégués ont également eu une séance pratique et utile sur la collecte d’échantillons d’ADN en utilisant les kits médicaux-légaux spéciaux disponibles.

62ème réunion du Comité permanent de la CITESLe chargé scientifique au GSRAf a participé à la 62ème réunion du Comité permanent de la CITES et a pris part à la rédaction d’un bref document du Groupe de travail de la CITES sur le rhinocéros. Le chargé scientifique et le président ont aussi fourni un rapport actualisé qui a été imprimé par l’UICN et mis à la disposition des délégués. Le Secrétariat du GSRAf a également fait des commentaires sur un projet de rapport du Groupe de travail de la CITES sur le rhinocéros pour la CdP 16 de la CITES qui avait été produit juste avant la date limite de soumission par le président sortant du Groupe de travail. Malheureusement, les préoccupations exprimées par le GSRAf en ce qui concerne la baisse de motivation des propriétaires du secteur privé et des communautés pour conserver les rhinocéros et l’augmentation inquiétante du désinvestissement des propriétaires dans la conservation du rhinocéros n’ont pas été capturées dans la version finale soumise au Secrétariat de la CITES. Le rapport souligne la nécessité de réduire la demande. J’espère qu’il y aura des opportunités pour plus de débats approfondis lors des prochaines réunions du Groupe de travail, ainsi qu’entre les sessions.

D’autres conférences

Au cours de la période considérée, le président et le chargé scientifique du GSRAf ont fait des présentations, et faisaient partie d’un groupe de discussion sur les défis et les approches de la conservation des rhinocéros au Congrès International sur la Gestion de la Faune à Durban en juillet 2012. En septembre 2012 le président a également participé et a pris la parole lors de la session du Groupe Consultatif Technique sur le Rhinocéros à la conférence annuelle de l’Association européenne des zoos et des aquariums à



A number of other rhino-related papers were presented at the conference. In November, the scientific officer also gave an invited presentation at the annual conference of the Professional Hunters Association of South Africa.

Global Environment Facility project developmentA number of South African AfRSG members met with the consultant who is helping South Africa’s Department of Environmental Affairs apply for a Global Environment Facility project in the hope of securing a possible USD 2.69 million in donor funds to boost wildlife forensic capabilities and the analysis of wildlife crime intelligence information in South Africa. A number of AfRSG members will be assisting by providing direct inputs into the final preparation of the application document.

AcknowledgementsI once again would like to acknowledge and thank our various sponsors: WWF’s African Rhino Programme (with funding from WWF- Netherlands), US Fish and Wildlife’s Rhino and Tiger Conservation Fund, Save the Rhino International (SRI), International Rhino Foundation and UK’s Department for Environment, Food and Rural Affairs for their sponsorship of the scientific officer’s time and next year’s AfRSG meeting. The Endangered Wildlife Trust is also thanked for administrative assistance. The assistance of SRI’s Cathy Dean and Susie Offord with fund raising, reporting and logistical support in helping organize the next AfRSG meeting is appreciated. I also thank Dr Richard Emslie, scientific officer, and Mr Ben Okita-Ouma, deputy Chair, for their inputs, constant support and advice.

Innsbruck en Autriche. La situation précaire à laquelle le rhinocéros d’Afrique est confronté y a été présentée. On a également fait un appel pour une plus grande coopération entre les zoos et les communautés de la conservation, surtout pour soutenir des programmes clés de conservation des rhinocéros sur le terrain. Le chargé scientifique a également prononcé un discours lors d’un symposium sur les pratiques contemporaines de conservation qui s’est tenu dans le KwaZulu-Natal en Afrique du Sud en octobre 2012. Il y avait d’autres présentations se rapportant au rhinocéros lors de la conférence. En novembre, le chargé scientifique a également donné une présentation à la conférence annuelle de l’Association des Chasseurs Professionnels de l’Afrique du Sud.

Elaboration du projet du Fonds pour l’Environnement MondialPlusieurs de membres du GSRAf d’Afrique du Sud se sont réunis avec le consultant qui aide le Ministère sud-africain des affaires environnementales à élaborer un projet pour le Fonds pour l’Environnement Mondial (FEM) dans l’espoir d’obtenir un financement possible de $ 2,69m des bailleurs de fonds pour renforcer les capacités médico-légales en matière de la faune et l’analyse des renseignements relatifs au crime de la faune en Afrique du Sud. Des membres du GSRAf donneront des contributions directes au document final qui est en cours de préparation.

RemerciementsJe voudrais encore une fois remercier nos différents sponsors: le Programme du WWF pour le rhinocéros d’Afrique (avec un financement du WWF Pays-Bas), le Fonds pour la Conservation du Rhinocéros et du Tigre du Service de la Pêche et de la Faune des Etats-Unis, Save the Rhino International, la Fondation Internationale pour le Rhinocéros et le DEFRA du Royaume-Uni pour leur soutien du chargé scientifique et de la réunion du GSRAf l’année prochaine. Je remercie également le Fonds pour la Faune et la Flore menacées d’extinction pour l’assistance administrative. L’aide de Cathy Dean et Susie Offord de SRI pour la collecte de fonds, la rédaction du rapport et la logistique pour l’organisation de la prochaine réunion du GSRAf est appréciée. Je remercie également le Dr Richard Emslie (chargé scientifique) et Mr. Ben Okita (vice-président) pour leur contribution, leur soutien constant et leurs conseils.


Talukdar

IUCN World Conservation Congress in Jeju, KoreaThe Chair of AsRSG attended the IUCN World Conservation Congress in Jeju, Republic of Korea, 5–16 September 2012. In the meeting of specialist groups organized by the Species Survival Commission of IUCN in Jeju on 6 September, the AsRSG Chair delivered a presentation on activities AsRSG was carrying out and on the current status of Asian rhinos in Asia. During the IUCN Congress, side meetings were held among the SSC Chair, Indonesian official delegates and IUCN members actively engaged in rhino conservation in Asia to discuss ways and means to further promote the conservation of the critically endangered Javan and Sumatran rhinos in Southeast Asia.

Joint report for CITES CoP16AsRSG, in close association with AfRSG and TRAFFIC, has drafted the joint report on rhinos for the upcoming CITES CoP16 scheduled for Bangkok in March 2013. The report can be viewed on the CITES website at http://www.cites.org/eng/cop/16/doc/E-CoP16-54-02.pdf.

Indian Rhino Vision 2020Between 2008 and 2012, about 18 wild rhinos were captured from Pabitora Wildlife Sanctuary and Kaziranga National Park (NP) of Assam and translocated to Manas NP under the Indian Rhino Vision 2020, a joint programme of the Assam Forest Department, Bodoland Territorial Council, the International Rhino Foundation, the US Fish and Wildlife Service and WWF-India. Two

Asian Rhino Specialist Group reportRapport du Groupe de Spécialistes des Rhinocéros d’Asie

Bibhab K. Talukdar, Chair/Président

Aaranyak, 50 Samanwoy Path (Survey), PO Beltola, Guwahati – 781 028, Assam, Indiaemail: [email protected]

Le Congrès Mondial de l’UICN sur la Conservation à Jeju, en Corée Le Président du GSRAs a participé au Congrès mondial de l’UICN sur la Conservation à Jeju, en Corée, du 5 au 16 septembre 2012. Lors de la réunion des groupes de spécialistes organisée par la Commission pour la Survie des Espèces de l’UICN à Jeju le 6 Septembre, le Président du GSRAs a présenté un exposé sur les activités que le GSRAs menait ainsi que l’état actuel des rhinocéros en Asie. Lors du Congrès de l’UICN, des réunions parallèles ont eu lieu entre le Président de la CSE, les délégués officiels indonésiens et les membres de l’UICN activement engagés dans la conservation des rhinocéros en Asie pour discuter les voies et moyens de promouvoir davantage la conservation du rhinocéros de Java et de Sumatra qui sont en danger critique d’extinction en Asie du Sud.

Rapport conjoint de la CITES pour la CdP16 Le GSRAs, en association étroite avec le GSRAf et TRAFFIC, a rédigé le rapport conjoint sur les rhinocéros pour la prochaine CdP16 de la CITES prévue pour Bangkok en mars 2013. Le rapport peut être consulté sur le site Internet de la CITES sur http://www.cites.org/eng/cop/16/doc/E-CoP16-54-02.pdf.

Vision 2020 pour le rhinocéros d’IndeEntre 2008 et 2012, environ 18 rhinocéros sauvages ont été capturés dans le Sanctuaire de la Faune de Pabitora et au Parc National de Kaziranga de l’Assam et transportés jusqu’au Parc national de Manas dans le cadre de la Vision 2020 pour le rhinocéros d’Inde, un programme conjoint du Département des Forêts de l’Assam, le Conseil territorial de Bodoland, la Fondation Internationale pour le Rhinocéros, le Service de la Pêche et de la Faune des


Asian Rhino Specialist Group report

translocated rhinos were poached in Manas NP as of 31 December 2012; one female translocated rhino gave birth to a calf at the end of September 2012.

The state of the greater one-horned rhino in South AsiaWithin India, 2012 witnessed more than 20 greater one-horned rhinos being poached in Assam, most of them in and around Kaziranga NP. Rhinos were poached in all the existing rhino-bearing areas of Assam—Pabitora Wildlife Sanctuary, Orang NP, Kaziranga NP and Manas NP—reflecting that poachers and rhino horn traders are putting their energies to target rhinos everywhere. During the last part of September, in the span of a week poachers taking advantage of the seasonal flood killed about five rhinos in Kaziranga NP. This in turn prompted the Honourable Gauhati High Court to take a suo motto public interest litigation on rhino poaching [where a government agency acts on its own cognizance – Ed.], directing the state government to file affidavits in the court.

Nepal has fought back rhino poachers strongly and could keep its rhino population protected from poachers, although one rhino was reported killed by poachers during 2012.

The state of Sumatran and Javan rhinos in Southeast AsiaThe current state of the Javan rhinoceros, which is restricted to the western part of Java, Indonesia, is almost static and its current population could be as low as 34 or as high as 44 rhinos, based on a recent camera trapping exercise carried out by authorities of Ujung Kulon NP. However, the promising news is that the rhinos are breeding and producing calves, and the rhino protection units of Ujung Kulon NP are recording them through footprints.

The current status of the Sumatran rhinoceros in Indonesia seems stable, but there is no evidence of its existence in Penisular Malaysia, although the government of Mayalsia has initiated a survey. The Sumatran rhino population in Sabah, Malaysia, is reported to be isolated, and the future of these rhinos cannot be assured. Consequently,

Etats-Unis et WWF-Inde. Deux rhinocéros transférés ont fait l’objet du braconnage au Parc national de Manas avant le 31 décembre 2012; un rhinocéros femelle transféré a donné naissance à un bébé rhinocéros à la fin de septembre 2012.

L’état du grand rhinocéros unicorne en Asie du SudEn Inde, plus de 20 grands rhinocéros unicornes ont fait l’objet du braconnage dans l’Assam en 2012, la plupart d’entre eux au Parc national de Kaziranga et ses alentours. Les rhinocéros ont été braconnés dans toutes les aires existantes de l’Assam abritant les rhinocéros – le Sanctuaire de la Faune de Pabitora, le Parc national d’Orang, le Parc national de Kaziranga et le Parc national de Manas – ce qui reflète le fait que les braconniers et les négociants des cornes de rhinocéros font l’effort de cibler les rhinocéros partout. Au cours de la dernière partie de septembre, en l’espace de quelques semaines les braconniers ont profité des inondations saisonnières pour tuer environ cinq rhinocéros dans le Parc national de Kaziranga. Cela à son tour a obligé la Haute Cour de Gauhati d’entreprendre de sa propre initiative un procès d’intérêt public suo motto sur le braconnage des rhinocéros [où un organisme gouvernemental agit de sa propre initiative – éd.], obligeant le gouvernement de l’Etat à déposer des affidavits dans la cour.

Le Népal s’est fortement battu contre les braconniers et a pu protéger sa population de rhinocéros, même si on rapporte qu’un rhinocéros a été tué par des braconniers en 2012.

L’état des rhinocéros de Sumatra et de Java en Asie du SudL’état actuel du rhinocéros de Java, qui est limité à la partie occidentale de Java, en Indonésie, est presque statique et sa population actuelle pourrait être aussi bas que 34 rhinocéros ou aussi haut que 44 rhinocéros, en se basant sur un exercice de piégeage photographique récente mené par les autorités du Parc national d’Ujung Kulon. Cependant, la bonne nouvelle est que les rhinocéros sont en train de se reproduire, et les unités de protection des rhinocéros du Parc national d’Ujung Kulon les enregistrent par le biais de leurs empreintes. L’état actuel du rhinocéros de Sumatra en Indonésie semble stable, mais il n’existe aucune preuve de son existence dans la péninsule de Malaisie, même si le Gouvernement


Talukdar

the Borneo Rhino Alliance (BORA) and the government of Sabah are planning to capture the isolated rhinos and put them in larger breeding areas within forests to save the species. Under the leadership of SSC Chair Dr Simon Stuart assisted by officials of BORA, a Sumatran Rhino Conservation Summit is being planned for 1–4 April 2013 in Singapore to discuss issues related to the Sumatran rhino.

de la Malaisie a organisé un recensement. On rapporte que la population des rhinocéros de Sumatra à Sabah, en Malaisie, est isolée, et l’avenir de ces rhinocéros ne peut être assuré. Par conséquent, l’Alliance du Rhinocéros de Bornéo (BORA) et le Gouvernement de Sabah envisagent de capturer ces rhinocéros isolés et les mettre dans de plus larges aires de reproduction dans les forêts pour sauver l’espèce. Sous la direction du Président de la CSE, le Dr. Stuart Simon, aidé par les représentants de BORA, on prévoit d’organiser un sommet sur la Conservation du Rhinocéros de Sumatra du 1 au 4 avril 2013 à Singapour pour discuter des questions liées au rhinocéros de Sumatra.


Functional relationship between crop raiding and habitat variables

Functional relationship between crop raiding by the savannah elephant and habitat variables of the Red Volta Valley in north-eastern Ghana

Patrick Adjewodah,1* William Oduro2 and Alex Asase3

1 Ghana Ecological Research Centre, PO Box KN 6365, Kaneshie-Accra, Ghana *corresponding author email: [email protected] and [email protected] College of Agriculture and Natural Resources, Kwame Nkrumah University of Science and Technology, University Post Office, Kumasi, Ghana3 Department of Botany, University of Ghana, PO Box LG 55, Legon, Ghana

AbstractWe investigated the degree to which crop raiding by elephants in the Red Volta Valley is a function of: the density and diversity of their natural browse, extent of degradation of their habitat, and proximity of crop enclaves to the nearest forest boundary. We assessed 50 m x 5 m quadrats for browse and for agents of degradation (m2 of quadrat area clear-felled, burned, and surface mined), and we adopted records of crop-raiding rates in 2000–2002 for correlation with habitat variables. We estimated the distance of affected enclaves from the forest, river and village using GIS; the total size of crop fields in a locale provided an estimate of the extent of cultivation. Using correlation and regression analysis, we examined for each locale the association between rate of crop-raiding incidents and the density and diversity of browse, size of crop fields, and distance of affected enclaves from the forest, river and village. We used a stepwise regression model to establish a functional relationship between crop raiding and the enumerated habitat variables. Second-order jackknife and Michaelis-Menten asymptotic estimators showed that the density and diversity of browse was near optimum. About 99% of the vegetation area sampled was burned, and 0.35 ha of vegetation was clear-felled for firewood. No evidence of mining was recorded. There was a significant inverse association between crop raiding and distance of affected enclaves from the forest, while the association between crop raiding and density and diversity of browse and the extent of degradation were not significant. A stepwise regression model gave a functional relationship between the observed rate of crop-raiding incidents and proximity of crop enclaves to the forest as Y = 25.105 + 3.2 – 9.73X (Y = rate of crop-raiding incidents and X the distance from enclave to the forest boundary). Contrary to speculation, crop raiding in the study area is not influenced by the status of browse, extent of degradation or size of cultivated fields. Thus, mitigation measures should include relocating farms away from forest reserves, while measures aimed at reducing the density of farms or replanting of the forest should be de-emphasized as they will not necessarily reduce crop raiding.

Additional key words: locale, enclave

RésuméNous avons étudié la mesure dans laquelle la maraude des cultures par les éléphants dans la vallée de la Volta Rouge dépend de: la densité et la diversité de leur broutage naturel, l’ampleur de la dégradation de leur habitat et la proximité des enclaves de cultures à la limite de la forêt la plus proche. Nous avons évalué des

RESEARCH


Adjewodah et al.

Introduction

The extent of natural browse—expressed in terms of density, diversity and structure of woody vegetation—is one of the key habitat requirements by elephants (McShane 1987; Sukumar 1990, 2003). A large proportion of the elephant’s natural diet is from browse. Browse forms over 50% of their diet (Bell 1985) and elephants need to take in 4–6% of their weight in browse each day. Barnes et al. (1995) observed a positive correlation between elephant crop raiding and habitat status and reported that increased crop-raiding levels are directly related to habitat degradation. In localities where natural vegetation is cleared for crops or burned, fewer browse resources are available (Okoumassou et al. 1998; Sam et al. 1998) and elephants are thought to seek alternative sources of food within crop fields (Sukumar 2003).

The spread of agriculture is a major source of habitat degradation (Hoare & Du Toit 1999). As human population grows, the need for more farmland has caused further degradation of elephant habitat and shifting of crop fields further into areas previously

occupied by elephants (Sam et al. 1998; Hoare & Du Toit 1999; Sukumar 2003). As a result incidents of elephant crop raiding have intensified where crop fields are widespread and where enclaves of cultivation have shifted closer into the elephants’ range (Sukumar 1990, 2003; Oppong et al. 2008).

Uncontrolled and recurrent incidents of wildfire are said to pose a great constraint on habitat and resource availability to wildlife in some savannah ecosystems (Oteng-Yeboah & Asase 2001). In the Red Volta Valley, fire is not controlled or used as a management tool as prescribed for wildlife protected areas in the climaxed savannahs (McShane 1987; Spinage 1994). Burning vegetation is common even at the latter part of the dry season (Ayigsi 1997; NCRC 1999, 2000) when fires are most intense and thus kill rather than support the growth of seedlings in the recruitment class (Oteng-Yeboah & Asase 2001). This type of burning could have an adverse impact on browse and cover for elephants.

Artisanal mining in the Red Volta Valley is common in the dry season. Two methods are commonly used: pit (or surface) mining, and alluvial mining. In the former, local artisans use hoes and other hand-held

quadrats de 50m x 5m pour le broutage et les agents de dégradation (m2 de superficie de quadrats coupés à blanc, brûlés, et la surface exploitée par les mines), et nous avons adopté des données des taux de maraude des cultures de 2000 à 2002 pour la corrélation avec les variables d’habitat. Nous avons estimé la distance des enclaves affectées à partir de la forêt, de la rivière et du village en utilisant le SIG. La superficie totale des champs cultivés dans un endroit a donné une estimation de l’étendue des cultures. En utilisant la corrélation et l’analyse de régression, nous avons examiné pour chaque endroit l’association entre le taux d’incidents de maraude des cultures et la densité et la diversité du broutage, la superficie des champs cultivés, et la distance des enclaves affectées à partir de la forêt, de la rivière et du village. Nous avons utilisé un modèle de régression par étapes pour établir une relation fonctionnelle entre la maraude des cultures et les variables énumérés de l’habitat. Des estimateurs asymptotiques de second ordre de Jackknife et de Michaelis-Menten ont montré que la densité et la diversité du broutage étaient quasi optimales. Environ 99% de la zone couverte de végétation échantillonnée avait été brûlée, et 0,35ha de végétation avait été coupée à blanc pour le bois de chauffage. Aucune preuve de l’exploitation minière n’a été enregistrée. Il y avait une association inverse significative entre la maraude des cultures et la distance des enclaves affectées à partir de la forêt, tandis que l’association entre la maraude des cultures, la densité et la diversité du broutage, et l’ampleur de la dégradation, n’étaient pas significatifs. Un modèle de régression par étapes a donné une relation fonctionnelle entre le taux observé d’incidents de maraude des cultures et la proximité des enclaves cultivées de la forêt de Y = 25.105 + 3.2 – 9.73X (Y = taux d’incidents de maraude des cultures, et X la distance de l’enclave à la limite de la forêt). Contrairement aux spéculations, la maraude des cultures dans la zone d’étude n’est pas influencée par l’état du broutage, l’ampleur de la dégradation ou la superficie des champs cultivés. Ainsi, les mesures d’atténuation devraient inclure la délocalisation des fermes loin des réserves forestières, tandis que les mesures visant à réduire la densité des exploitations agricoles ou la replantation de la forêt sont moins importantes car elles ne réduiront pas nécessairement la maraude des cultures.

Mots clés supplémentaires: endroit, enclave



implements to dig up the soil and process it for gold ore. In the process vegetation cover is removed and several pits filled with water are left behind. Alluvial mining involves collecting alluvial silt from the riverbed. The practice is sporadic along sections of the river during the dry season.

Elephants in the Red Volta Valley use the forest reserves as a natural refuge and raid crops in fields adjacent to the reserves (Jachmann 1992; Stalmans & Anderson 1992; Sam 1994; Okoumassou et al. 1998). However, knowledge of the key variables that influence the rate of crop-raiding incidents is scanty and speculative. Previous workers have suggested that the rate of elephant crop-raiding incidents could be associated with degradation of the elephant’s natural habitat and to dwindling browse resources (Ayigsi 1997; Okoumassou et al. 1998; Sam et al. 1998; Adjewodah et al. 2003; Adjewodah 2004). Considering the socio-economic effects of crop damage on the local communities and their implications for the conservation of elephants in the Red Volta Valley, a rigorous analysis of the factors that influence elephant crop raiding in the area is necessary.

In this paper, we examine the association between elephant crop raiding and some key variables in the Red Volta Valley of Ghana. The objectives of the study were to 1) identify from a set of potential cause factors variables that might influence local elephant crop raiding and 2) develop a cost-effective statistical model for determining the rate of elephant crop-raiding incidents.

Study areaThe Red Volta Valley comprises a network of adjoin-ing forest reserves—Red Volta East, Red Volta West, Gambaga Scarp East, Gambaga Scarp West, and Morago East Forest Reserves—and adjacent off-reserve woodlands, fallow land and crop fields (latitude 10°30’ to 11°00’ N, longitude 0°45’ to 0°00’ W). The Red Volta Valley falls within Talensi–Nabdam, Bawku West and Bongo Districts of the Upper East Region of Ghana (Fig. 1). The vegetation is savannah woodland and consists of deciduous short trees and shrubs (Oteng-Yeboah 2001; Barnes et al. 2006b). The banks of the Red Volta, White Volta and Morago rivers are lined with gallery forest.

The forest reserves harbour transboundary elephant migratory routes that link the Red Volta Valley to Kabore–Tambi National Park and Nazinga Game

Ranch (in south central Burkina Faso), and to Fosse-aux-Lions National Park in northern Togo (Sebogo & Barnes 2003; Barnes et al. 2006a). About 1,049 km2 of the Red Volta Valley comprising the above forest reserves and adjoining fallow woodlands is uninhabited and thus potentially available for elephants.

The Forestry Services Division of the Forestry Commission manages the Red Volta forest reserves to protect watersheds. Wildlife management is not a major priority. 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. Mean annual rainfall is about 896 mm with an annual peak in July, August and September. The dry period is characterized by desiccating north-east winds known as harmattan, which bring dust and haze from the Sahara Desert. The dry season is also characterized by a high incidence of wildfires between December and February, and during this period wildfire damage to vegetation extends over a large portion of the study area (Adjewodah 2004).

The main economic activity of the people is rainfed subsistence agriculture. Farm sizes are small and range from 0.1 ha to 7.3 ha (Adjewodah et al. 2005). Crops common to the area include millet, maize and groundnut. The Red Volta Valley has been discussed in further detail elsewhere (Sam et al. 1998; Adjewodah et al. 2005).

Materials and methodsWith the help of a topographical map, we divided the uninhabited natural area of the Red Volta Valley into 18 locales. Each of the locales measured 7.63 km x 7.63 km and encompassed the gallery forest along the Red Volta, White Volta and Morago rivers, and adjoining savannah fallow lands. We randomly selected six locales—Bongo, Buing, Kusanaba, Morago, Sakote and Tilli—as the focal areas of the study. The selected locales were put into grids at 5-minute intervals (on both the longitudinal and latitudinal axes), resulting in sub-units referred to as cells (Fig. 2). A locale consisted of about 18 cells each measuring 1.8 km x 1.8 km. We randomly selected 10 cells from each locale. We delineated a 1,000-m-long transect within each cell. We determined the terminal coordinates of each transect from a field map beforehand, and field groups used a GPS, a compass and tape measure to follow the transect line. Using the tape measure, we measured 2.5 m on both sides


Adjewodah et al.

of the transect line to create 50 m x 5 m segments (quadrats) along each transect. There were 20 adjoining segments or quadrats along each transect. We surveyed 52 transects instead of 60, as 8 of the selected transects fell in inaccessible terrain and were not surveyed.

To quantitatively describe habitat available to and used by elephants, we enumerated the vegetation in 4 of the 20 quadrats, that is, in every 5th segment along the 1000-m transect (Oteng-Yeboah 2001). We tallied trees above diameter at breast height (dbh) by species and diameter classes, and tallied shrubs over 30 cm tall by species. Trees and shrubs selectively browsed by elephants of the neighbouring Kabore-Tambi National Park (Spinage 1985) provided the basis for identifying browse species among the enumerated plants. KAKUM

0 100 km

ANKABA

DADICAO

GOABO

CHICHEDOM DIGYA

MOLE

RED VOLTSISALI-TUMU

NANDOM

Zebilla Natinga

Widinaba

TilliNangodi

Bongo

BURKINA FAS0

TOGOMorago East Forest Reserve

Sakote

Datoko

Kusanaba

Zongoiri

Gambaga

Nalerigo

Sakogu

Nakpanduri

Bimbago

Buipielise

Bugwia

Garu Natinga

Buing/Degare

Shiega

WHITEVOLTA

WHITE V OLTA

RED

VOLTA

WH

ITE VOLTA

10º 45’

11º 00’

00º 00’00º 15’00º 30’00º 45’

00º 45’ 00º 30’10º 30’

00º 00’SCALE 1:320,000

roadsriversinternational boundarytowns/villagesforestTounga farmssampled farms

N

Gambaga Scarp EastForest Reserve

Morago WestForest Reserve

Gambaga Scarp WestForest Reserve

Red Volta WestForest Reserve

Red Volta West Forest Reserve

Red V

olta

West F

ore

st Rese

rve

Red Volta EastForest Reserve

Figure 1. Location of the Red Volta Valley in north-eastern Ghana and the study area.

BURKINA FASO

TOGO

White Volta river

Morago river

Red Volta river

White Volta river

10º 50’

10º 40’

00º 30’ 0’ E00º 40’ 0’ E 0º 20’ 0’ E 0º 10’ 0’ E

10º 30’

0º 10’ 0’ E0º 20’ 0’ E

0 3.75 7.5 15 22.5 30 km

0º 30’ 0’ E0º 40’ 0’ E

10º 30’ 0’ N

10º 40’ 0’ N

10º 50’ 0’ N

LegendLocale

Boundary transect

Rivers

Forest reserveboundary

N

Figure 2. The Red Volta ecosystem showing the research locales and transects.



For extent of degradation, we measured the cross-sectional area (in m2) of mining pits within each of the selected 50 m x 5 m vegetation quadrats and took size measurements (in m2) of the area of the quadrat clear-felled for firewood and the area (in m2) affected by wildfire. We obtained distance measurements of enclaves of cultivation to the nearest forest boundary by estimating the mean distance of a random sample of farms from the reserve boundary. We also measured the distances of enclaves of cultivation to the nearest village (human settlement) and to the Red Volta River using GIS. We adopted the size of crop field recorded and rate of elephant crop-raiding incidents reported in 2000–2002 (Adjewodah et al. 2005) for correlation with the extent of browse and the other habitat variables.

We estimated browse using the density of woody stem and Michaelis-Menten species diversity measurements (Gotelli & Colwell 2001; Sukumar 2003). We used the statistical programme EstimateS (Colwell 2005) for rigorous extrapolation and interpolations of the empirical sample-based dataset, and computed sample-based rarefaction curves of the diversity and density of browse by repeated resampling (Gotelli & Colwell 2001). We calculated the second-order jackknife and Michaelis-Menten species richness and diversity indices for each locale. ANOVA was used to compare the density of browse and non-browse species among locales, and the independent t-test to compare density of browse species with the density of all other enumerated species. Species diversity indices including observed species richness, second-order jackknife, and Michaelis-Menten estimators (Gotelli & Colwell 2001) provided a test of homogeneity or otherwise of the habitat in the six locales in relation to distribution and diversity of woody plants.

We used Pearson moment product correlation test (Zar 1999) to determine the association between browse density and the rate of elephant crop-raiding incidents. Standard linear regression was used to determine the type and strength of the relationship between rate of elephant crop-raiding incidents and distance of affected enclaves from the nearest forest boundary. Spearman correlation was used to estimate the association between the rate of incidents and size of crop field.

We used Statistical Programme for Social Sciences (SPSS Inc. 2007) to build a model of the rate of crop-raiding incidents following the Stepwise Strategy (Dytham 2003). The statistical package identified

the key factor(s) of rate of crop raiding (dependent variable) in a locale from a set of habitat variables: 1) size of crop field cultivated in the locale, 2) density of woody stem, 3) density of browse, 4) distance of enclave from the forest boundary, 5) distance of enclave from the Red Volta River, and 6) distance of the locale’s enclave from settlement. The analysis produced a function between crop raiding and the selected cause factor(s). We added and subtracted the variables listed above in steps using only those combinations and slopes that generated a better fit (Dytham 2003). This allowed a stepwise algorithm to choose the most important factor(s) and to select a ‘best’ model for predicting the rate of elephant crop raiding in the study area. We used ANOVA to test the acceptability of the model from a statistical perspective, and used a histogram and normality plot of the residuals to check the assumption of normality of the error term. We based the model on the equation below (Zar 1999): Yj = a + b1X1j + b2X2j + b3X3j + …. + bmXmj + ej where Yj = number of incidents in the jth locale a = the intercept (the value of Y when X1, X2 and

X3 are zero) m = number of predictors (or independent variables) b1 = correlation coefficient of the 1st variable X1j = denotes the jth observation of variable X1 ej = is the error in the observed value for the jth case

ResultsA total of 12,948 trees and shrubs were enumerated during the study. About 16% (n = 67) of the species recorded were categorized as elephant browse based on a list of browse species compiled by Spinage (1985) (Table 1). All the browse species were present in each of the locales, except for Borassus aethiopum, which was recorded only in Sakote (Table 1).

The density of browse species was not significantly different between locales (ANOVA: F5,66 = 0.52, P = 0.774, homogeneity of variances, P = 0.487). When the density of all enumerated plants was compared (Table 2), ANOVA indicates that neither was the density of plants in general significantly different between the locales (F5,46 = 0.44, P = 0.818).

To further explore the data for any variations among locales, the density of browse species was compared with the density of all other enumerated species to determine relative abundance (Table 3). In this instance


Adjewodah et al.

Table 3. Number of stems per m2 of both browse and other woody plants in six locales in the Red Volta Valley

Locale

Density (no. of stems/m2)Non-browse woody

plants (range)Non-browse woody

plants (mean)Browse (mean)

Bongo 0.09–0.38 0.266 0.137Buing 0.05–0.40 0.268 0.134Kusanaba 0.15–0.33 0.229 0.137Morago 0.14–0.39 0.233 0.079Sakote 0.06–0.45 0.272 0.105Tilli 0.06–0.41 0.222 0.080

independent t-test indicates there is a significant difference between browse and non-browse species, and thus affirmed the dominance of the former group of plants (Levene’s test of equality of variances: P > 0.05; t-test, df =10, P < 0.05).

Observed species richness (Sob ), second-order jackknife, and Michaelis-Menten estimators were computed (Table 4) and compared for differences in diversity of species. The Kruskal-Wallis test indicates homogeneity of the habitat in relation to plant diversity (species richness: chi-square = 5, P > 0.05; second-order jackknife: P > 0.05; Michaelis-Menten: chi-square =

5, P > 0.05).The observed species diversity (richness) within

locales was compared with the Michaelis-Menten asymptotic estimate. An independent sample t-test showed no significant difference between the observed species richness (Sob ) and the projected Michaelis-Menten asymptote (Levene’s test of equality of variances: P > 0.05, t-test for equality of means P = 0.198). Species rarefaction curves (based on a series of 100 randomizations) showed close similarity of the vegetation within the locales (Fig. 3).

Table 2. Density of woody plants enumerated in six localities of the Red Volta Valley

LocaleDensity (no. of stems/m2)

T1 T2 T3 T4 T5 T6 T7 T8 T9 T10

Bongo 0.381 0.296 0.235 0.340 0.357 0.251 0.153 0.293 0.092 –Buing 0.195 0.195 0.263 0.400 0.367 0.289 0.326 0.325 0.268 0.052Kusanaba 0.327 0.244 0.336 0.257 0.233 0.187 0.131 0.233 0.188 0.152Morago 0.389 0.290 0.188 0.174 0.085 0.362 0.143 – – –Sakote 0.328 0.450 0.340 0.060 0.340 0.115 – – – –Tilli 0.227 0.249 0.205 0.334 0.408 0.061 0.202 0.238 0.138 0.162

Ty = transect y; – = missing data

Table 1. Density and distribution of elephant browse in the Red Volta Valley

Browse species Density (no. of stems/m2)

Bongo Buing Kusanaba Morago Sakote TilliAcacia hockii 0.288 0.042 0.252 0.093 0.207 0.093Acacia nilotica 0.372 0.002 0.218 0.013 0.464 0.013Borassus aethiopum 0.000 0.000 0.000 0.000 0.001 0.000Combretum sp. 0.264 0.129 0.473 0.231 0.126 0.231Detarium microcarpum 0.057 0.652 0.070 0.032 0.006 0.032Gardenia ternifolia 0.082 0.032 0.013 0.017 0.010 0.017Lannea acida 0.011 0.004 0.006 0.005 0.002 0.005Mitragyna sp. 0.028 0.021 0.001 0.000 0.00 0.013Piliostigma thonningii 0.050 0.018 0.117 0.042 0.019 0.042Pteleopsis suberosa 0.155 0.351 0.137 0.156 0.142 0.156Terminalia macroptera 0.235 0.218 0.196 0.288 0.198 0.288Vitellaria paradoxa 0.098 0.136 0.161 0.073 0.087 0.073



Table 4. Composition and diversity estimates for 208 vegetation quadrats within the six locales (standard deviations are in brackets)

Locale

Observed species richness (Sob) 1 Density (SD) 2

Michaelis-Menten estimate 3

2nd-order jackknife estimate 4

Singletone (SD) 5

Doubletone (SD) 6

Bongo 32 0.38 (0.09) 33.38 37.42 1.71 (1.06) 2.35 (0.90)Buing 37 0.20 (0.10) 42.04 48.03 4.15 (2.15) 3.49 (2.10)Kusanaba 42 0.23 (0.07) 48.37 50.16 4.01 (2.34) 4.52 (1.88)Morago 48 0.23 (0.12) 58.14 49.51 3.05 (3.37) 4.01 (3.57)Sakote 39 0.27 (0.15) 47.61 42.51 4.64 (2.4) 4.60 (2.11)Tilli 34 0.23 (0.09) 37.92 40.65 1.84 (1.48) 2.76 (1.68)

1 Sob Observed species richness estimated as number of species expected in the pooled samples given the empirical data2 Stem density (of all diameter classes) per square metre3 Michaelis-Menten estimate of species richness: an asymptote estimator4 Second-order jackknife richness estimator (mean among runs)5 Singletone mean: number of species with only one individual in the pooled quadrats6 Doubletone mean: number of species with only two individuals in the pooled data

plants were within size class 15–25 cm and 4% were greater than 35 cm in girth size (Table 6).

In this analysis, density of woody stem and species diversity (weighed by the Michaelis-Menten species richness) provided an estimate of browse, and quadrat area burned or clear-felled provided an estimate of degradation (Table 7).

Pearson product–moment correlation indicates a weak association between the rate of incidents and density (Pearson: r = 0.806, df = 5, P > 0.05), diversity (Pearson: r = –0.007, df = 5, P = 0.98) of browse. Similarly, there was no significant association between rate of incidents and quadrat area burned (Pearson product moment correlation: r = –0.557, df = 5, P = 0.329), and clear-felled (Pearson product moment correlation: r = –0.016, df = 5, P = 0.979). The extent of cultivation in a locale is estimated by the size of crop field (Table 7). When the analysis was run, Spearman correlation indicated a positive but weak association between size of cropland cultivated and rate of crop-raiding incidents (r = 0.25, df = 5, P = 0.63).

The mean distance of a sample of farms from the nearest forest reserve boundary provides an estimate for proximity of an enclave to the forest reserve (core elephant habitat) (Table 8). Correlation analysis shows a significant association between crop raiding and proximity of fields to the forest reserve. The number of crop-raiding incidents markedly increases with decreasing distance to the forest reserve (Pearson product moment correlation: r = –0.96, df = 5, P = 0.002) (Fig. 4).

Correlation analysis indicates an inverse association between the rate of crop-raiding incidents and distance

Extent of habitat degradation in the Red Volta Valley

Evidence of wildfire was recorded in most of the quadrats sampled, and about 99% of the vegetation area sampled was severely burned (Table 5). There was no evidence of mining in any of the vegetation plots, and 0.35 ha involving 14 quadrats (n = 208) was clear-felled for firewood (Table 5).

The vegetation was dominated (73–82%) by small plants with girth size less than 5 cm diameter (Table 6), typical of a fire pro-climax (savannah) habitat. Plants within size 25–35 cm were the least common, making up only 3% of total plants enumerated; 5% of

Buing Sakote

Bongo Kusanaba Tilli Morago

Spec

ies

Individuals

Figure 3. The mean species accumulation curve (observed species richness) of woody stems in six locales of the Red Volta Valley elephant range.


Adjewodah et al.

of crop fields from the Red Volta River (Pearson product moment correlation: r = –0.43, df = 5, P = 0.39), but the association was not statistically significant. In contrast, the association between the rate of crop-raiding incidents and distance of the nearest village to the affected fields is not significant (r = –0.213, P = 0.68) (Fig. 5).

Functional relationship between crop raiding and habitat variablesA functional relationship between the rate of crop-raiding incidents and some key habitat variables was determined using a regression model. Eleven

potential variables of the rate of crop raiding are considered in this analysis (Table 9).

Stepwise algorithm chooses ‘distance of the affected farmed enclave to the forest boundary’ as the only predictor of rate of elephant crop raiding in a locale. Equation 1 provides the relationship between the rate of crop-raiding incidents and proximity of farmed enclave to the forest.

Y = 25.105 + 3.2 – 9.73X (Equation 1)where

Y = expected rate of incidentsX = distance of the affected enclave from the

nearest forest boundary

Table 5. Extent of habitat degradation observed in six locales

Locale

Vegetation burned by fire Vegetation clear-felled

No. of quadrats burned

Area (m2) % of area sample

No. of quadrats Area (m2)

BongoBuingKusanabaMoragoSakoteTilli

363436282432

9,0008,5009,0007,0006,0008,000

100100100100100100

038210

0750

2,000500250

0

Table 6. Distribution of vegetation among five diameter classes in six locales within the Red Volta Valley

LocaleDiameter class in cm (%) Number of

< 5 5–15 15–25 25–35 > 35 Plants Transects Trees/ha


827379738176

712 8161010

376547

443224

444433

2,4002,6842,2941,6961,6352,239

91010 7 610

804724481654518537

Table 7. Rate of incident, extent of cultivation, extent of browse and habitat degradation in six locales of the Red Volta Valley during 2000–2002

Locale

No. of elephant

crop-raiding incidents

Extent of cultivationa

(ha)

Density and diversity of browse Extent of habitat degradation

Mean density of browse species

Mean density of all species

Diversity index (Michaelis-

Menten)

Quadrat area burned

(ha)

Quadrat area clear-felled

(ha)


99

186

453

250.6585.2221.9462.7

1,104.6250.6

0.1370.1340.1370.0790.1050.080

0.2660.2680.2290.2330.2720.222

33.3842.0448.3758.1447.6137.92

08,5009,0007,0006,0008,000

0750

2,000500250

0a Adjewodah et al. 2005



Rate of incidents = Intercept + standard error of estimate + coefficient x distance of enclave from forest boundary (Equation 2)

The model, equation 1, is arrived at when the standard error estimate and coefficients of the regression (Table 10) are accounted for in equation 2. It indicates an inverse relationship between rate of crop-raiding incidents and distance of the affected enclave from the forest boundary, and predicts a best guess of the rate of incidents in a locality of the Red Volta to average 4.83 ± 2.75 crop-raiding incidents per annum. The model states that the expected rate of crop-raiding incidents in a locale is equal to 25.1 x distance of enclave from the forest minus 9.73. For Sakote, where distance of enclave to forest boundary is 1.27 km, the predicted rate of crop-raiding incidents using this model is 15.95 incidents per annum.

Table 8. Proximity of cultivated enclaves to the nearest forest reserve boundary, the Red Volta River, and settlements, and the annual frequency of elephant crop-raiding in the Red Volta range during 2000–2002

Locale Mean distance of farm enclave or affected farms (km) Crop-raiding incidents (no.)from main

villageto Red Volta

Valleyfrom forest boundary

2000 2001 2002 Mean

BongoBiungKusanabaMoragoSakoteTilli

1.284.367.104.241.582.62

5.247.998.26

14.454.815.33

2.92.502.233.001.272.65

0820

280

20

12651

7140

122

3362

151

Figure 4. Relationship between distance of cultivated enclaves from the forest, and annual rate of elephant crop raiding in Red Volta Valley in 2000–2002.

1.000

10

20

30R2 linear = 0.04

1.50 2.00Distance of farm from forest boundary (km)

Num

ber o

f cro

p-ra

idin

g in

cide

nts

2.50 3.00

7

10

5

20

15

25

30

Num

ber o

f cro

p-ra

idin

g in

cide

nts

2

R2 linear = 4.677E–4

4Distance of farm enclave from village (km)

6 8

Figure 5. Relationship between number of crop-raiding incidents and distance of cultivated enclaves from the nearest village.

The ratio of the regression and residual sums of squares indicates that 70% of the variation in crop-raiding incidents is explained by the model (regression sum of squares = 98.79, residual = 42.04, total = 140.8; R2 = 0.72). Furthermore, ANOVA indicates that the variation explained by the model is not due to chance (F = 9.40, P = 0.037). The histogram and normality plot of the residuals fairly follows the shape of the normal curve (Fig. 6).

The multiple correlation coefficient (R) is examined for the strength of the relationship between the model-predicted values and the observed rate of crop-raiding incidents. When this was carried out, stepwise regression indicated a high correlation coefficient (R = 0.838), and a strong coefficient of determination (R2 = 0.702). As a further measure of the strength of the model fit, the standard error (2.24 incidents per annum) compares favourably with the standard deviation of incidents (of 2.75 incidents per annum).


Adjewodah et al.

DiscussionThe Red Volta Valley holds adequate browse resources for elephants; 2 out of every 10 woody plants are edible for them. In the savannah of Uganda, examination of the stomach contents of elephants revealed only 25 species in their diet, which compares with the 11 browse species recorded in the Red Volta Valley. Woody stems were homogeneously distributed among locales, and the abundance, diversity and richness of browse were near the asymptotic estimates or optimum values expected for the area.

Evidence of damage to the vegetation by wildfire was widespread. Though an assessment of the impact of fire on elephant browse was beyond the scope of this study, observations from elsewhere in Ghana (Oteng-Yeboah & Asase 2001) indicate that uncontrolled fire can constrain the availability of resources to wildlife. For instance, it can be assumed from the results that only 20–30% (plants that are 5 cm dbh or larger) of the enumerated plants in the Red Volta are actually available to elephants, as the animals browse from 1 to 2 m above ground level (Spinage 1994). Plants in the recruitment stage (< 5 cm dbh), which form a majority of the specimens enumerated, fall short of the preferred browse height and elephants would not select them.

The results however indicate there is no marked association among crop raiding, the extent of browse and habitat degradation. Though some level of association was found between the rate of crop-raiding incidents and the extent of cultivation within a locale, this association was weak and lacked statistical significance. In parallel, Sukumar (2003) found no clear relationship between crop raiding and the status of elephant habitat in the immediate vicinity of cultivated fields, and noted that crop raiding by elephants was influenced by several attributes of their habitat which were not mutually exclusive, but interrelated in a complex fashion that was poorly understood.

Regression standardized residual

Freq

uenc

y

–1.0 –0.5 0.0 0.5 1.00

2

4

6

8

10Dependent variable: mean no. of incidentsMean = –4.56E–16SD = 0.392n =14

Figure 6. Histogram and normality plot of the residuals against rate of crop-raiding incidents.

Tabl

e 9.

Pot

entia

l pre

dict

or v

aria

bles

of t

he ra

te o

f ele

phan

t cro

p-ra

idin

g in

cide

nts

in s

ix lo

calit

ies

of th

e Re

d Vo

lta V

alle

y

Loca

le

Cro

p ar

ea

de st

roye

d (h

a)

Woo

dy p

lant

de

nsity

(n

o./m

2 )Di

vers

ity

inde

xRa

ids

(no.

/yr)

Farm

ers

(no.

)Ar

ea c

ulti-

vate

d (h

a)

Dist

ance

of e

ncla

ve (k

m)

Area

bur

ned

(ha)

Area

cle

ared

(h

a)fro

m

villa

geto

rive

rfro

m

fore

st

Bong

o 5.

438

0.26

633

.38

311

425

11.

285.

2–

––

Buin

g0.

60.

268

42.0

43

195

585

4.36

8.0

–8,

500

750

Kusa

naba

3.

184

0.22

948

.37

682

222

7.10

8.3

2.33

9,00

02,

000

Mor

ago

2.65

30.

233

58.1

41

160

463

4.24

14.0

–7,

000

500

Sako

te

5.87

80.

272

47.6

115

398

1,10

51.

584.

81.

276,

000

250

Tilli

0.63

40.

222

37.9

21

112

251

2.62

5.3

2.65

8,00

00

– =

mis

sing

dat

a



Barnes et al. (2003) found that the rate of crop raiding was proportional to the density of farms. Similarly, Sukumar (1990) found that the frequency of crop raiding during different months of the year was proportional to the area of land under cultivation. However, they made these observations at conflict sites where elephants are present year round. In the Red Volta Valley, the elephant population is seasonal and the extent and rate of crop raiding is more dependent on the migration of elephants into the study area across the frontier with Burkina Faso (Adjewodah 2004). In 2003, no crop-raiding incident was reported even though the size of cultivated land and crop availability had not changed markedly from previous seasons when incidents were high (Adjewodah et al. 2005). This was attributed to a marked change in the elephant migration pattern (Adjewodah 2004), as elephants from Burkina Faso did not arrive in the study area during the crop-raiding season (Sawadogo 2003).

A marked association between the rate of crop-raiding incidents and proximity of enclaves to the forest was observed. This confirms suggestions by previous workers (Sam et al. 1997; NCRC 1999, 2000) that elephant preference for forest reserves means farms nearer the forest are at greater risk of being affected. This preference thus provides scientific evidence to support suggestions that relocating farms away from the forest reserve will result in a reduced rate of crop-raiding incidents in the Red Volta Valley. Stepwise algorithm regression analysis selected ‘distance of affected enclave to the forest boundary’ as the variable that best accounts for the rate of crop-raiding incidents in a locale. The other potential variables were ignored by the stepwise algorithm because after adding ‘distance of enclave to the forest’ none of them made any significant addition to the model. Taking Sakote, for instance, where the distance of the locale’s farmed enclave is only 1.27 km from the forest boundary, the predicted rate of crop-raiding incidents generated by the model (15.948 incidents per annum) compares closely with

the observed rate, which is 15 incidents per annum. The model does a good job of modelling the rate of incidents in a locale as the R2 value of 0.72 means that the model explains about 70% of the variation in rate of incident. The significant F statistic further indicates that the prediction of rate of incidents using the model is statistically reliable.

ConclusionThe mention of inadequate browse resources as an underlying cause of elephant crop raiding in the Red Volta Valley is not upheld by this study. The Red Volta Valley holds adequate browse resources for elephants and the diversity of woody plants is near the optimum expected value for the area. Evidence of bush burning was extensive and poses the greatest threat to elephant browse resources relative to pit mining and clear-felling of vegetation. The high percentage of enumerated plants within the recruitment size is a manifestation of the effect of fire on vegetation.

The rate of crop loss can be predicted from proximity of cultivation to the nearest forest boundary but is not markedly influenced by the extent of degradation and the diversity of woody plants. Thus a mitigation plan targeted at relocating fields away from the forest reserves will yield a positive outcome, while measures aimed at reducing the density of farms or replanting of the forest with browse species may not necessarily reduce the rate and extent of elephant crop raiding in the study area.

AcknowledgementsThis research was funded by the European Commis-sion through the Small Grants Programme of the African Elephant Specialist Group (AfESG) of the Species Survival Commission (SSC), World Conservation Union (IUCN). Earthwatch Institute also provided support for follow-up work in the

Table 10. Coefficients of regression and collinearity statistics of regression model

Model 1 Unstandardized coefficients

Standardized coefficients

T Sig. Correlations Collinearity statistics

B Std. error Beta Zero-order Partial Part Tolerance VIF(Constant) 25.105 6.743 3.723 0.020Distance of farms from forest boundary

–9.730 3.174 –0.838 –3.066 0.037 –0.838 –0.838 –0.838 1.000 1.000


Adjewodah et al.

Red Volta Valley in collaboration with the Nature Conservation Research Centre. The kind support of these institutions is deeply appreciated.

We thank the following individuals who in many ways gave us the needed support through their advice and guidance. Prof. Paul Beier of Northern Arizona University visited with the lead author during the fieldwork and critiqued the methodology. We are grateful to the field staff for their endurance and hard work.

ReferencesAdjewodah, P. 2004. Habitat status, population and

distribution of the African savannah elephant (Loxodonta africana) in north-eastern Ghana. IUCN/AfESG Project SG0203, Nairobi. Unpublished project report.

Adjewodah, P., Murphy, A., and Mason, J. 2003. Mitigating elephant crop raiding: The Red Volta Valley experience, Ghana. Available http://iucn.org/themes/ssc/sgs/afesg/hec/pdfs/hecrvvrep.pdf.

Adjewodah, P., Beier, P., Sam, M.K., and Mason, J.J. 2005. Elephant crop damage in the Red Volta Valley, north-eastern Ghana. Pachyderm 38:39–48.

Ayigsi, J. 1997. Volta–Morago community elephant reserve project: summary of field report. Nature Conservation Research Centre, Accra. Unpublished.

Barnes, R.F.W., Adjewodah, P., Ouedraogo, L., Godwin, P., Seynou, O., Tiendrebeogo, C. and Pousga Célestin Zida, C.P. 2006a. A route map for the preparation of management plans for the transfrontier ecosystem between Mali, Burkina Faso and Ghana. Project report. IUCN/AfESG, Burkina Faso. 54 pp. Unpublished.

Barnes, R.F.W., Adjewodah, P., Ouedraogo, L., Héma, E., Ouiminga, H. and Pousga Célestin Zida, C.P. 2006b. Transfrontier corridors for West African elephants: The PONASI–Red Volta and Sahelian Corridor Project. IUCN/AfESG, Burkina Faso. 50 pp. Unpublished.

Barnes, R.F.W., Azika, S. and Asamoah-Boateng, B. 1995. Timber, cocoa and crop-raiding elephants: a preliminary study from southern Ghana. Pachyderm 19:33–38.

Barnes, R.F.W., Boafo, Y., Nandjui, A., Dubiure, U., Hema, E.M., Danquah, E. and Manford, M. 2003. An overview of crop raiding by elephants around the Kakum Conservation Area: Part 2. Technical report. Elephant Biology and Management Project, Africa Program, Conservation International. Unpublished.

Bell, R.H.V. 1985. Elephants and woodland—a reply. Pachyderm 5:17–18.

Colwell, R.K. 2005. Estimates: statistical estimation of species richness and shared species from samples. Version 7.5. User’s guide and application. Published at http://purl.oclc.org/estimates.

Dytham, C. 2003. Choosing and using statistics: a biologist’s guide. Oxford, UK: Blackwell Science.

Gotelli, N.J. and Colwell, R.K. 2001. Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecological Letters 4:379–391.

Hoare, R. and Du Toit, J.T. 1999. Coexistence between people and elephants in the African savannahs. Conservation Biology 13:633–639.

Jachmann, H. 1992. Movement of elephants in and around the Nazinga Game Ranch, Burkina Faso. Journal of African Zoology 106:27–37.

McShane, T.O. 1987. Elephant–fire relationships in combretum/terminalia woodland in Southwest Niger. African Journal of Ecology 25:79–94.

[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. Project report 1999. Nature Conservation Research Centre/Technoserve, Accra, Ghana. Unpublished.

[NCRC] Nature Conservation Research Centre. 2000. The micro-best practices project: monitoring report on the ecological impact of the project for the second period. TechnoServe Inc., Accra, Ghana. Unpublished.

Okoumassou, K., Barnes, R.F.W. and Sam, M. 1998. The distribution of elephants in north-eastern Ghana and northern Togo. Pachyderm 26:52–60.

Oppong, S.K., Danquah, E. and Sam, M.K. 2008. An update on crop raiding by elephants at Bia Conservation Area, Ghana, from 2004 to 2006. Pachyderm 44:59–64.

Oteng-Yeboah, A.A. and Asase, A. 2001. Wechiau Community Hippo Sanctuary: Earthwatch Itineraries, June 2000–February 2001. Botanical Surveys. Nature Conservation Research Centre, Accra. Unpublished.

Sam, M.K. 1994. A preliminary survey of elephants in Northern Ghana. Project report 1994. Ghana Wildlife Department, Accra Ghana. Unpublished.

Sam, M.K., Barnes, R.F.W. and Okoumassou, K. 1998. Elephants, human ecology and environmental degradation in north-eastern Ghana and northern Togo. Pachyderm 26:61–68.



Sam, M.K., Haizel, C. and Barnes, R.F.W. 1997. Crop-raiding by elephants during the 1996 harvest season in the Red Volta Valley (Upper East Region, Ghana). Project report, 1997. Wildlife Department, Accra, Ghana. Unpublished.

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. Project report. IUCN/AfESG, Nairobi, Kenya. Unpublished.

Sebogo, L. and Barnes, R.F.W. (eds.). 2003. Action plan for the management of transfrontier elephant conservation corridors in West Africa. Technical Report, 2003, IUCN/AfESG Ouagadougou, 49 pp. Unpublished.

Spinage, C.A. 1985. The elephants of Burkina Faso, West Africa. Pachyderm 5:2–5.

Spinage, C.A. 1994. Elephants. London: T and AD Poyser Ltd.

SPSS Inc. 2007. Statistical programme for data analyses. USA: Polar Engineering and Consultancy.

Stalmans, M. and Anderson, J. 1992. The forest elephants of Togo. African Wildlife 46:71–75.

Sukumar, R. 1990. Ecology of the Asian elephant in southern India. II. Feeding habits and raiding patterns. Journal of Tropical Ecology 6:33–53.

Sukumar, R. 2003. The living elephants: evolutionary ecology, behaviour, and conservation. New York: Oxford University Press.

Zar, J.H. 1999. Biostatistical analysis. Upper Saddle River, NJ: Prentice Hall.


Aristide et al.

Indentification et caractérisation des formations végétales exploitées par l’éléphant Loxodonta africana dans la Réserve de Biosphère de la Pendjari au Nord-Ouest de la République du BéninTehou C. Aristide,¹,2,* Kossou Eric,² Mensah G. Apolinaire,3 Houinato Marcel 2 and Sinsin Augustin Brice 2

¹ Projet W–Arly–Pendjari, 02 B.P. 527, Cotonou 01, République du Bénin * [email protected]² Laboratoire d’Ecologie Appliquée/FSA/UAC, 01 B.P. 526 Recette Principale, Cotonou 01, République du Bénin3 Institut National des Recherches Agricoles du Bénin, 01 B.P. 2359, Recette Principale, Cotonou 01, République du Bénin

Résumé Les formations végétales de la Réserve de Biosphère de la Pendjari (RBP) sont soumises à une utilisation permanente par la population des éléphants. Ces formations représentent une importante source d’alimentation pour la faune en général et en particulier pour les éléphants mais aussi une source d’approvisionnement en viande de gibier et autres produits forestiers non ligneux pour les populations riveraines. Peu de données scientifiques existent sur les relations entre les formations végétales exploitées par les éléphants et les éléphants dans la RBP. Les objectifs de l’étude sont les suivants: identifier les différentes zones exploitées par les éléphants ; caractériser les différentes formations végétales exploitées par les éléphants ; et évaluer les dégâts occasionnés par les éléphants sur ces formations. La méthode de Braun-Blanquet a été utilisée pour les relevés phytosociologiques. Les matrices des données collectées à partir des relevés ont été traitées avec le logiciel STATISTICA. Le dendrogramme a été obtenu à l’aide du logiciel STATISTICA par la méthode de Ward sur la base des distances euclidiennes pour définir les groupements végétaux. La surface terrière, la structure verticale et la structure horizontale ont été les paramètres dendrométriques calculés. L’ensemble des données a été soumis au test de Monté Carlo pour analyser les corrélations entre les facteurs environnementaux et les différentes phytocénoses exploitées par les éléphants dans le but d’évaluer l’impact des éléphants sur les formations végétales. Les résultats ont montré que les zones de Porga et d’Arly étaient des zones de forte concentration en éléphants, tandis que celles de Batia et Konkombri étaient des zones de faible concentration en éléphants. Sur les 61 relevés x 183 espèces effectués sur l’ensemble des zones identifiées, le dendrogramme a fait ressortir 5 groupements végétaux discriminés suivants les facteurs type de formations végétales, niveau de dégradation due à la densité en éléphants et la superficie des zones abritant ces groupements. La structure verticale observée était une distribution s’ajustant dans l’ensemble à une distribution en cloche de Gauss. La structure horizontale était en J renversé et traduisant une formation naturelle. Selon la typologie des dégâts occasionnés par les éléphants, la présence des chablis étaient fortement corrélés avec les zones de forte concentration en éléphants. En somme, la caractérisation dendrométrique faite souligne que la végétation de la RBP demeure en bon état de conservation malgré son caractère giboyeux dans la sous région de l’Afrique de l’ouest.

Mots clés supplémentaires : phytosociologie, dendrométrie, Chablis, régénération

Abstract The plant communities of the Pendjari Biosphere Reserve (PBR) are continuously consumed by the elephant population. These plant communities are an important source of food for wildlife in general and for elephants in particular, and they are also a source of bushmeat and other non-woody forest products for the neighbouring populations. Little scientific data exist on the relationship between the plant species that elephants eat and


Formation végétales exploitées par l’éléphant dans la Réserve de Biosphère de la Pendjari

elephants in the PBR. The objectives of the study were to: identify the different zones elephants use, characterize the different plant communities elephants consume, and evaluate elephant damage on the plant communities. The Braun-Blanquet method was used for the phytosociological survey. Matrices of data collected from the survey were processed using STATISTICA software. The dendrogram was obtained using STATISTICA software by Ward’s method using Euclidian distances to define plant species. The earth’s surface, the vertical and horizontal structures, were the dendrometric parameters calculated. All data were subjected to the Monte Carlo test in order to analyse the correlations between environmental factors and the different phytocenoses elephants use to evaluate their impact on plant communities. The results showed that Porga and Arly zones had high concentrations of elephants while Batia and Konkombri had low concentrations. Out of 61 plants surveyed x 183 species in all the zones identified, the dendrogram identified five groups of plants differentiated according to typical characteristics of plant communities, the level of degradation caused by elephants and the area of the zones containing these groups of plants. The vertical structure observed corresponded mainly with a Gaussian bell-shaped distribution. The horizontal structure was an inverted J and resulted from a natural formation. As regards the type of damage caused by elephants, the presence of felled trees was strongly correlated with areas of high concentration of elephants. Overall, the dendrometric characterization carried out underlines the fact that the vegetation in the PBR is still in a good state of conservation despite the number of wild animals in this West African sub-region.

Additional key words: phytosociology, dendrometrics, felled trees, regeneration

Introduction

La population d’éléphants de l’ordre de 1808 têtes de la Réserve de Biosphère de la Pendjari (RBP) constitue la plus importante de l’ensemble des aires protégées du Bénin (Sinsin et al. 2006). Ce taxon constitue un maillon important dans le processus de la régénération des communautés végétales dans les savanes en Afrique. De ce fait, cette espèce mérite une attention particulière pour capitaliser les efforts inhérents à sa conservation. Il est nécessaire de prendre en compte cette espèce dans l’étude de la régénération des forêts et des savanes en Afrique qui a été jusqu’ici une préoccupation majeure des spécialistes de la gestion de la faune. A l’instar d’autres espèces fauniques, la dynamique des populations d’éléphants est affectée par divers facteurs écologiques et anthropiques.

Les types de formations végétales de la zone soudano-sahélienne du Nord-Bénin constituent une ressource alimentaire importante pour la faune sauvage, voire aussi un habitat pour certaines espèces animales comme des mammifères rongeurs, oiseaux, reptiles, etc. En effet, l’étude de l’écologie et de l’éthologie de la faune sauvage des aires protégées du Bénin a permis de disposer des connaissances sur l’interaction végétation et animaux sauvages comme c’est le cas des éléphants (Tèhou 2001).

L’éléphant fait preuve d’une souplesse écologique stupéfiante (Pfeiffer 1989). En tant qu’un strict

herbivore sauvage, il présente une gamme de structure variée et de mœurs dans les savanes des aires protégées où il ingère principalement des graminées, des feuilles d’arbre ou de buissons et des fruits. La présence des éléphants dans les aires protégées et savanes est plutôt bénéfique et contribue à l’équilibre de la flore et de la faune. En général, les éléphants consomment au niveau de la strate arborescente des gousses de légumineuses, des baies, des drupes et des samares. Les différentes parties consommées au niveau des espèces de cette strate sont les rameaux, les feuilles, les fruits, les écorces et les racines. Au niveau de la strate herbacée, les éléphants broutent aussi les graminées. Dans les exploitations agricoles, le régime alimentaire est surtout constitué de noix de karité lorsque cette espèce est en fructification. Entre le mois de mai et de juillet, d’autres espèces consommées dans les champs sont Sclerocarya birrea, Parkia biglobosa, Mangifera indica, Zea mays, Dioscorea alata, Cajanus cajan, Sorghum bicolor, Vigna unguiculata, Arachis hypogaea et Gossypium hirsutum (Tèhou et Sinsin 1999). Une bonne connaissance des écosystèmes devient de plus en plus indispensable pour une gestion communautaire durable des populations d’éléphants (Tèhou 2001). Tout ce qui précède justifie la présente étude sur l’identification et la caractérisation des formations végétales exploitées par l’éléphant Loxodonta africana dans la Réserve de Biosphère de la Pendjari.


Aristide et al.

Milieu d’étude

L’étude a lieu dans la Réserve de Biosphère de Pendjari située géographiquement au 10°30’ et 11°30’N, 0°50’ et 2°00’E. Ce milieu bénéficie d’un climat de type soudanien et une pluviométrie unimodale de 1.100 mm.

Matériels et méthodes

Matériels

La carte de végétation de la réserve a été utilisée pour connaître toutes les formations végétales de la RBP et pour les identifier. Les cartes annuelles des mouvements des éléphants dans la RBP ont été utilisées pour la superposition des différentes années afin d’avoir la distribution des éléphants dans le temps et dans l’espace. La carte des différentes bandes de précipitations de la RBP a permis d’avoir des informations sur les mouvements des éléphants dans les différentes strates en rapport avec l’intensité des pluies. Un GPS a été utilisé pour géoréférencier les données collectées sur le terrain. Les couloirs de passage, les crottes, les espèces appétées et les dégâts occasionnés par les éléphants et autres ont été mentionnés dans un carnet de notes.

La caractérisation floristique des formations végétales (Sokpon et al. 2001) parcourues par l’éléphant dans la RBP a été faite avec la mise en place des placeaux de 50m x 20m à l’aide d’un ruban de 50m, d’une boussole de marque SUNTO pour l´orthogonalité des angles aux sommets des placeaux, des piquets de coins en bois coupés à l’aide d’un coupe-coupe et enfoncés à l’aide d’un marteau pour matérialiser les placeaux. Des fiches ont été élaborées pour faire des relevés phytosociologiques et dendrométriques. Pour la réalisation d’un herbier, un sécateur de jardinier a été utilisé pour prélever les échantillons de plantes transportés dans des sachets et sacs poubelles en plastique, rangés, conservés et protégés dans des sachets de productivité en papier buvard. Un clinomètre a été utilisé pour la mesure des pentes et des hauteurs. Un ruban pii a été utilisé pour la mesure des circonférences.

MéthodesL’identification des zones de répartition des éléphants dans la RBP a été effectuée en les deux phases suivantes : une phase d’analyse bibliographique et de

laboratoire afin de vérifier les données collectées sur le terrain ; et une phase de collecte de données dans la zone d’étude. A l’aide du logiciel Arcview GIS3.2 la carte de végétation a été superposée à la carte de présence des éléphants dans la RBP ; les deux cartes étant de même échelle.

Pour la caractérisation floristique des formations végétales exploitées par les éléphants dans la RBP, les placeaux d’inventaire ont été installés dans les formations répondant aux critères de sélection et en tenant aussi bien compte de la taille de la formation dans la réserve, de la topographie, de l’accessibilité, de la présence ou non d’eau, du type de sol et de l’homogénéité floristique. Ainsi, les relevés phytosociologiques ont été effectués suivant la méthode sigmatiste de Braun-Blanquet (1932). Au niveau de chaque site, l’inventaire des espèces ligneuses a été réalisé à l’intérieur des placeaux de 20m x 50m (1.000 m2) et celui des herbacées était fait dans des placettes de 10m x 10m (100m2) installées à l’intérieur du placeau de 20m x 50m suivants les facies en présence dans le milieu (savanes, forêts claires et galeries forestières). A l’intérieur de ces placeaux la caractérisation floristique des formations végétales parcourues par l’éléphant dans la RBP a été faite. Les relevés dendrométriques ont été effectués dans les mêmes placeaux de 20m x 50m utilisés pour les relevés phytosociologiques. Le diamètre à hauteur d’homme (dbh) des individus d’arbres a été le paramètre collecté. Un transect a été installé perpendiculairement au lit mineur du cours d’eau considéré. Le nombre d’individus de chaque espèce a été déterminé. En dehors des relevés effectués au sein des placeaux, un relevé itinérant a été fait au niveau de tous les transects et il a contribué à l’estimation de la richesse spécifique de chaque zone (Yedomonhan 2002).

Analyse des donnéesLa matrice de données constituée de 61 relevés phytosociologiques et de 183 espèces végétales ont été soumises à l’aide du logiciel CANOCO (Canonical Community Ordination) version 4.5 à une DCA (Dentrended Correspondence Analysis) pour déterminer les groupements végétaux. La classification hiérarchique des relevés (clustering) a été réalisée sous ce même logiciel.

La richesse spécifique et la diversité d’une communauté végétale sont les deux éléments qui recouvrent la détermination de la richesse d’un



taxon. Ainsi, le nombre de familles a été dénombré (diversité de familles) de même que les espèces (richesse spécifique). La diversité spécifique a été analysée à partir de la richesse spécifique (S), l’indice de diversité de Shannon et le coefficient d’équitabilité de Pielou.

IndIce de dIversIté de shannon (h’ en bIts)La formule de l’indice de diversité de Shannon (H’ en bits) est la suivante (Shannon et Weiner 1949 dans Shannon et Weiner 1963) :

avec :

Pi = (ni/n) est la fréquence relative des individus de l’espèce (i) ;

(ni) est le nombre de fois que l’espèce (i) est contactée et (n) est l’effort total de prospection dans la station. L’indice de Shannon Weaner exprime donc la quantité d’information nécessaire à la description de l’avifaune d’un milieu.

equItabIlIté de PIelou (e)

L’équitabilité de Pielou (E) est souvent calculée pour traduire le degré de diversité atteint par rapport au maximum possible. Elle varie de 0 à 1 et son expression est la suivante :

,

avec : H’ max = log2(S), où S est le nombre total d’espèces dans la station considérée.

Les individus d’arbres dans chaque groupement ont été groupés en des classes de diamètre de 10 cm pour construire l’histogramme de la structure diamétrique de chaque groupement. La structure des groupements a été ajustée au modèle de Weibull à cause de sa grande flexibilité (Johnson et Kotz 1970 ; Bonou et al. 2009). La fonction de densité de probabilité de la distribution de Weibull est donnée par la formule

exp[-( ;

où : x est le diamètre de l’arbre ia est le paramètre de position (ici a = 10) ;b est le paramètre d’échelle ou de taille ; c est le paramètre de forme lié à la structure

observée. La caractérisation des peuplements est à faire sur base du coefficient de forme c (Bonou et al. 2009).

Résultats

IdentIfIcatIon des grouPements végétaux

Les relevés phytosociologiques effectués dans les formations végétales exploitées par les éléphants dans la RBP correspondaient à une matrice de 61 relevés et de 183 espèces. Cette matrice soumise à une classification hiérarchique ascendante suivant la méthode de Ward et de Person a permis d’obtenir un dendrogramme issu de cette analyse (Figure 1).

L’analyse hiérarchique ascendante a permis l’identification des cinq groupements végétaux dans les différentes formations exploitées par les éléphants dans la RBP suivants :• des groupements végétaux GV1 et GV2 de la zone

de Porga rencontrés dans une zone de forte concen-tration en éléphant.

• des groupements GV3, GV4 GV5 du complexe Arly, Batia et Konkombri. Ce complexe regroupe une zone de forte concentration (Arly) et deux zones de faible concentration (Batia et Konkombri). Les groupements de forte concentration ont été

caractéristiques des zones riches en mares.

facteurs détermInant la réPartItIon sPatIale des grouPements végétaux

L’analyse de la figure 2 présentant la projection dans un plan factoriel des relevés phytosociologiques, a révélé que, dans le plan factoriel 1-2, l’axe 1 qui représentait 10,37 % (tableau 1) de l’inertie totale opposait d’une manière générale les relevés de la zone de Porga [comportant le groupement à Andropogon gayanus et Acacia gourmaensis (R33, R29, R38, R32, …, R35 )] à celui des zones de Arly, Batia et Konkombri [comportant le groupement à Andropogon pseudapricus et à Anogeissus leiocarpus (R17, R16, R7, R13, R10, R59, ……. , R1)] décrivant un caractère d`exiguïté de l`espace à Porga et une forte densité entraînant la dégradation du milieu due à un fort taux de charge en éléphant.

En somme, les facteurs déterminant l’identification et la répartition spatiale des groupements végétaux ont été le gradient pédologique en relation avec le niveau de dégradation due à la pression et à la densité suivant


Aristide et al.

Figure 1. Dendrogramme des 61 relevés effectués dans les formations exploitées par l’éléphant dans la RBP. G = groupe; GV = groupement végétal

Arbre de 61 relevésMéth. de Ward

1-r Pearson

(Dcl

a./D

max

)*10

0

0

20

40

60

80

100

120

R33

R29

R28

R37

R32

R27

R38

R36

R35

R32

R30

R23

R22

R21

R34

R26

R24

R25

R55

R58

R42

R57

R43

R40

R18

R61

R53

R47 R5

R9

R6

R14 R3

R56

R54

R20

R60

R45

R51

R50

R48 R2

R17

R16

R15

R11

R19

R39 R4

R49

R44

R46

R41

R12

R52 R7

R8

R13

R10

R59 R1

G1 G2

GV1 GV2 GV3 GV4 GV5

Figure 2. Projection dans un plan factoriel des 61 relevés effectués dans les formations exploitées par les éléphants dans la RBP.

0

0

8040

80

Axis 1

Axis

2

R1

R2R3

R4

R5

R6

R7

R8

R9R10

R11

R12R13

R14R15R16

R17

R18R39

R40

R41

R42

R30

R43

R44

R45

R46

R47

R 4 8

R49

R50

R51R52

R53

R54

R 55

R56

R57

R58

R59

R60

R61

R19

R20

R25R26

R23

R21

R24

R22

R30

R35

R36R37

R38

R32

R27

R28

R32

R 2 9R33

R34

Tableau 1 : Valeurs propres et variances expliquées par les axes

Axes Valeurs propres

Variances Variances cumulées

1 10,95 17,66 17,662 4,72 7,67 25,293 3,33 3,75 29,04



l`axe 1 et le gradient de couverture des formations. Tous les deux gradients étaient en relation avec les zones de concentration en éléphant. Toutefois, les axes 1 et 2 ne contribuaient que pour 15,67% de l’inertie totale. Ceci dénotait d’une dispersion de l’information sur les axes factoriels.

structure horIzontale

Au niveau du groupe A constitué par le groupement à Andropogon gayanus et Acacia gourmaensis et celui à Andropogon pseudapricus et Vitellaria paradoxa, il n’existait pas de très grands arbres (figure 3). Toutefois, le fait que ces groupements se trouvent sur le parcours des éléphants nous a amené à dire que ce phénomène pouvait être dû à la pression comme les dégâts chablis des éléphants sur certains arbres de grand diamètre comme Adansonia digitata par exemple, puisque le groupe appartenait à une zone caractérisée par de fortes concentrations en éléphants.

Au niveau du groupe B constitué par le groupement à Andropogon tectorum et Balanites aegyptiaca et celui à Andropogon pseudapricus et Anogeissus leiocarpa, la structure horizontale traduisait par le fait que les arbres ayant un grand diamètre ont été épargnés (figure 3).

Au niveau du groupe C constitué par le groupement à Aneilema lanceolatum et Acacia goumaensis, l’effectif des arbres de diamètre supérieur à 30 cm était considérable (figure 3).

La lecture de la structure horizontale dans les deux zones a permis de constater que l’effectif des ligneux à Dbh compris entre 25 et 30 cm était moins élevé en zone de forte concentration comparativement aux ligneux de cette même classe dans la zone à faible concentration. Ceci pouvait s’expliquer par le fait que les éléphants exercent une pression (mutilation des baobabs) sur les arbres de cette classe dans la zone de forte concentration.

Dans la zone à faible concentration, le phénomène contraire s’observait pour illustrer la faible pression due peut-être à la faible densité dans la zone et à la diminution des facteurs favorisant l’émergence des éléphants.

structure vertIcale

La structure verticale a montré que le groupe 1 était constitué par le groupement à Andropogon gayanus et Acacia gourmaensis et celui à Andropogon pseudapricus et Vitellaria paradoxa (figure 4). Ce

groupe était caractérisé par l’inexistence des individus de petite taille de 1 à 4 m de hauteur ce qui devait être dû tant à l’effet de piétinement par les éléphants qu’à l’absence d’individus de taille variant entre 10 et 12 m due au fait que ces arbres peuvent être l’objet de chablis ou l’objet d’aliments pour les petits et les grands éléphants. En effet, l’éléphant dans cette zone de forte concentration casse l’arbre qui perd en hauteur. Ce phénomène observé pour les individus des classes comprises entre 10 et 12 m pourrait être dû au fait que les éléphants par leur prélèvement sur les individus de basse classe empêchent la croissance en hauteur de ces derniers.

Le groupe 2 constitué du groupement à Andropogon tectorum et Balanites aegyptiaca et celui à Andropogon pseudapricus et Anogeissus leiocarpa, était caractérisé par la présence à la fois des individus de petite taille et de grande taille (figure 4). Cet état de choses était dû au fait que ce groupe se retrouve dans une zone de faible et de forte concentration en éléphants avec assez de conditions favorables n’ayant pas forcément de lien avec les ligneux comme la présence des mares et de grande superficies.

Le groupe 3 constitué par le groupement à Aneilema lanceolatum et Acacia goumaensis était caractérisé par l’absence des arbres de faible taille et la présence des arbres de grande taille (figure 4). Ceci était dû au fait que ces arbres ont été épargnés parce que dans le milieu la préférence alimentaire des éléphants ne se basait pas sur ces espèces ou parce que celles-ci sont peut-être toxiques.

De façon générale, la distribution en cloche de Gauss observée dans tous les groupements a montré l’impact des éléphants sur les arbres de faible hauteur par mutilation et par chablis et sur la régénération par piétinement (figure 4). Ainsi, une relation existait entre le diamètre et la hauteur des ligneux appétés par les éléphants puisque suivant le type de zone, l’importance des dégâts sur les ligneux s’exprimait d’abord par la mutilation et puis après la conséquence était que l’arbre perdait sa capacité de croître en hauteur ou était sujet à devenir un chablis plus tard.

Selon les deux graphes comparatifs illustrant la structure verticale dans les deux zones, l’effectif des individus de classe comprise entre 10 et 12 m de hauteur était faible dans la zone de forte concentration des éléphants. Ceci dénotait de la pression (chablis) enregistrée par les éléphants sur les espèces de cette classe. Cet état de choses n’était pas observé dans la zone de faible concentration des éléphants. Ainsi,


Aristide et al.

Figure 3. Structures horizontales des groupes A, B et C constitués par divers groupements végétaux.

Distribution par classe de diamètre du groupement à Andropogon gayanus et Acacia gourmaensis

y = –1,4124x3 + 19,976x2 - 92,171x + 140,34R2 = 0,99

01020304050607080

10–15 15–20 20–25 25–30 30– +

Classe de diamètre (cm)

E�ec

tif e

n (%

)

Distribution par classe de diamètre du groupement à Andropogon pseudapricus et à Vitellaria paradoxa

y = –5,8333x3 + 63,643x2 – 222,52x + 250R2 = 0,99

-100

20

40

60

80

100

10–15 15–20 20–25 25–30 30–+


E�ec

tif e

n (%

)

Structures horizontales du groupe A

Distribution par classe de diamètre du groupement à Andropogon tectorum et Balanites aegyptiaca

y = –2,6961x3 + 31,828x2 – 119,89x + 150,88R2 = 0,99

010203040506070

10–15 15–20 20–25 25–30 30–+


E�ec

tif e

n (%

)

Distribution par classe de diamètre du groupement à Andropogon pseudapricus et Anogeissus leiocarpa

y = –0,5587x3 + 8,9785x2 – 46,887x + 87,039R2 = 0,99

0

10

20

30

40

50

60

10–15 15–20 20–25 25–30 30– +Classe de diamètre (cm)

E�ec

tif e

n (%

)

Structures horizontales du groupe B

Distribution par classe de diamètre du groupement àAneilema lanceolatum et Acacia gourmaensis

y = 1,0511x3 - 3,0245x2 – 17,589x + 58,739R2 = 0,98

0

10

20

30

40

10–15 15–20 20–25 25–30 30–+Classe de diamètre (cm)

E�ec

tif e

n (%

)

Structures horizontales du groupe C

Distribution par classe de diamètre des ligneux de la zone de forte concentration

y = –1,7316x3 + 22,681x2 – 96367x + 137,53R2 = 0,99

0

10

20

30

40

50

60

70

10–15 15–20 20–25 25–30 30–+


E�ec

tif e

n (%

)

Distibution par classe de diamètre des ligneux de la zone de faible concentration

y = -3,5813x3 + 39,079x2 - 138,33x + 166,28R2 = 0,99

0

10

20

30

40

50

60

70

10–15 15–20 20–25 25–30 30–+Classe de diamètre (cm)

E�ec

tif e

n (%

)



Figure 4. Structures verticales des groupes 1, 2 et 3 constitués par divers groupements végétaux.

y = –0,2459x5 + 6,4642x4 - 60,805x3 + 247,88x2 – 413,29x + 220,84R2 = 0,78

-20-10

010203040506070

1–2 2–4 4–6 6–8 8–10 10–12 12–+

Classe dehauteur

Effe

ctif

en (%

)

y = -0,1389x5 + 4,3078x4 – 44,537x3 + 191,67x2 – 327,49x + 177,03R2 = 0,76

-20-10

010203040506070

1–2 2–4 4–6 6–8 8–10 10–12 12–+

Classe de hauteur

Effe

ctif

en(%

)

Structures verticales du groupe 1

y = 0,0714x5 – 1,0011x4 + 3,3983x3 + 3,5119x2 – 20,245x + 17,143R2 = 0,99

0

5

10

15

20

25

30

1–2 2–4 4–6 6–8 8–10 10–12 12– +Classe de hauteur

Effe

ctif

en (%

)

y = 0,0164x5 – 0,0112x4 – 2,9407x3 + 20,156x2 – 34,566x + 18,227R2 = 0,98

0

5

10

15

20

25

30

35

1–2 2–4 4–6 6–8 8–10 10–12 12–+Classe de hauteur

Effe

ctif

en

(%)

Structures verticales du groupe 2

y = –0,2825x5 + 6,2468x4 – 51,22x3 + 187,92x2 – 289,07x + 146,49R2 = 0,99

-10-505

101520253035

1–2 2–4 4–6 6–8 8–10 10–12 12–+

Classe de hauteur

Effe

ctif

en %

Distribution par classe de hauteur des ligneux en zone de forte concentration

y = -0,2392x5 + 6,0068x4 - 54,429x3 + 214,95x2 – 348,41x + 183,96R2 = 0,84

-20

-10

0

10

20

30

40

50

60

1–2 2–4 4–6 6–8 8–10 10–12 12–+

Classe de hauteur

Effe

ctif

en (%

)

Distribution par classe de hauteur des ligneuxen zone de faible concentration

y = 0,121x5 – 1,8267x4 + 8,1928x3 - 9,0726x2 – 1,8744x + 4,6083R2 = 0,97

-505

10152025303540

1–2 2–4 4–6 6–8 8–10 10–12 12–+

Classe des hauteurs

Effe

ctif

en (%

)

Distribution par classe de hauteur du groupement à Andropogon tectorum et Balanites aegyptiaca

Distribution par classe de hauteur du groupement à Andropogon pseudapricus et Anogeissus leiocarpa

Distribution par classe de hauteur du groupement àAneilema lanceolatum et Acacia gourmaensis

Structures verticales des groupements 3

Distribution par classe de hauteur du groupement àAndropogon pseudapricus et Vitellaria paradoxa

Distribution par classe de hauteur du groupement à Andropogon gayanus et Acacia gourmaensis


Aristide et al.

ces chablis pouvaient être causés lors du passage de l’éléphant ou au moment du broutage pour l’alimentation de ses éléphanteaux. De plus, ces zones étaient riches en espèces du genre acacia. Aussi, dans la zone de forte concentration des éléphants, la densité de ligneux est élevée et le taux de dégât était élevé.

Discussion

Formations végétales exploitées par les éléphantsDans la RBP, l’étude montre que les éléphants exploitent beaucoup plus les formations ouvertes caractérisées par la présence des mares et dans lesquelles ils trouvent assez d’espèces qu’ils appètent. Les formations plus ou moins fermées servent de lieu de repos mais plus tard deviennent ouvertes. Toutefois, les éléphants étant de bons disséminateurs (Alexandre 1978 ; Alfa Gambari 2002, 2003) de telles zones deviennent plus tard des zones très riches sur le plan de la diversité et de la richesse en espèces. Ce phénomène s’explique puisque par leurs crottes qu’ils laissent sur le lieu de repos, les éléphants participent bien à la régénération des espèces ce qui contribue à la reconstitution à long terme de la flore (White et al. 1993).

Au cours des investigations, les zones de forte concentration des éléphants à Arly et Porga présentent des caractéristiques telles que : abondance des mares, richesse en espèces appétées avec un niveau de protection très appréciable. Ces caractéristiques sont des indicateurs qui favorisent la présence des éléphants dans ces zones. Concernant l’abondance des mares où la concentration des éléphants est plus élevée, ce résultat est conforme à ceux de Laws (1993) et Ekobo (1995) cités par Mubalama (2000) qui ont noté des liens étroits entre les éléphants et les habitats « inondés » (marécages, bas-fonds et les forêts périodiquement inondées). De même, ce résultat est conforme à celui de Tehou et de Sinsin (1999) qui ont trouvé que les éléphants sont surtout localisés autour des mares et sur les affleurements.

Caractéristiques phytoécologiques des groupements végétaux

Le Parc National de la Pendjari fait partie d`un réseau d’aires protégées dans la sous-région de l’Afrique de l’Ouest et le résultat relatif à la richesse en espèces

appétées, est analogue aux observations faites par Kidjo (1992a, 1992b). D’ailleurs, Tehou (1995) lie la distribution spatiale des éléphants à la recherche alimentaire. En effet, du côté de Porga, en plus de ces caractéristiques, nous pouvons ajouter l’anthropisation car de ce côté, il y a la zone d’occupation contrôlée (ZOC) où se mènent des activités champêtres. Ainsi, parlant des activités champêtres, Tehou et Sinsin (1999) ont montré que le mouvement des éléphants vers les terroirs villageois est dû à un problème de préférence alimentaire basé surtout sur la phénologie des essences fruitières. Toutefois, dans cette étude la zone de Porga s’est distinguée du complexe Arly, Batia, Konkombri. Cet état de choses peut également être dû au fait que les formations végétales du complexe Arly, Batia et Konkombri sont dans un premier temps beaucoup plus riches en peuplement d’Anogeissus leiocarpa, de B. aegyptiaca, d’Acacia et d’autres espèces appétées par les éléphants. Par conséquent, à l’effet de la superficie s’ajoute l’effet de la composition floristique des formations végétales. C’est ce qui explique leur fréquentation dans les habitats purs d’Anogeissus leiocarpa, de B. aegyptiaca et d’Acacia qui leur servent non seulement de garde-manger mais aussi et surtout leur offrent un cadre de repos. Dans un second temps, le caractère de couverture de cette zone comparativement à celui de Porga est que le complexe Arly-Batia-Konkombri est composé de formations végétales plus ou moins fermées. Ceci se justifie parce que les éléphants affectionnent les formations ouvertes ou plus ou moins fermées (Tehou 2001). De plus, la richesse spécifique indiquée par les divers indices de diversité calculés montre que les zones pâturées par les éléphants ont une diversité en espèces végétales entrant dans leur alimentation considérable.

Le nombre de 183 espèces relevées dans les formations végétales exploitées par les éléphants lors de la présente étude est du même ordre que ceux obtenus par Legba (2005) avec 170 espèces recensées sur les collines de RBP, Gaoué (2000) avec 175 espèces inventoriées dans la plaine de la zone cynégétique de la Pendjari, et Avohou (2003) avec 185 espèces dans la région de Tanguieta-Batia sur la chaîne de l’Atacora. Par contre, ce nombre est faible par rapport aux 589 espèces enregistrées dans les Monts Kouffé au centre du Bénin par Houinato (2001). Ainsi, la faible richesse floristique est probablement due à la période de collecte des données en pleine saison des pluies où l’accès à plusieurs formations végétales ont été quasi impossible à cause de la montée des eaux.



Dans la RBP les familles les plus représentées sont similaires à celles recensées dans la région soudanienne par d’autres études comme celle de Houinato (2001) avec 99 familles au nombre desquelles les familles des Poaceae, des Fabaceae et des Asteraceae sont les plus représentées et celle de Wala (2004) sur la chaine de l’Atacora avec 107 familles dont les familles des Poaceae, des Fabaceae et des Rubiaceae sont les plus dominantes. Ceci peut s’expliquer par le fait que la RBP se retrouve en zones savanicoles où les espèces annuelles et pérennes dominent généralement la strate herbacée. De même, toutes les espèces caractérisant les ligneux de ces groupements végétaux sont des ligneux appétés par les éléphants.

Caractérisation dendrométrique

Dans la RBP les groupements végétaux étudiés sont caractérisés pour la plupart par des Phanérophytes en nombre important (41,4%). Houinato et Sinsin (2001) ont trouvé des proportions similaires pour les savanes humides et les forêts claires, en moyenne 36,3% pour les Phanérophytes et 34% pour les Thérophytes. Nos résultats dans ce domaine corroborent ceux trouvés par Gaoue (2000). L’abondance des Phanérophytes dans la quasi-totalité des groupements végétaux traduit le climat auquel appartient la zone d’étude. Aussi, cette prédominance des Phanérophytes peut être interprétée comme une bonne propension à la régénération des sols de la région. Les méga-Phanérophytes sont peu communes du fait que les savanes sont les principales formations végétales dans lesquelles les méso-Phanérophytes et les micro-Phanérophytes sont les plus fréquents (Avohou 2003). De plus, cette abondance explique la présence des éléphants car selon certains auteurs, les éléphants aiment plus les Phanérophytes de la classe méso et micro, certainement compte tenue de leur taille comprise entre 3 et 3,5 m. C’est ce qui justifie l’intégration des éléphants dans la Réserve de Biosphère de la Pendjari.

La physionomie des groupements végétaux étudiés est déterminée beaucoup plus par les Thérophytes et les Hémicryptophytes ce qui est le cas avec les résultats de Gaoué (2000) dans la plaine de la zone cynégétique de la Pendjari où tous les groupements observés sont essentiellement constitué de Thérophytes. Cela se justifie parce que les éléphants appètent beaucoup les Thérophytes (Tchamba 1996 ; Tchamba et Seme 1999). Aussi, cette dominance de Thérophytes peut constituer un indice de forte perturbation. Le fort taux

de Thérophytes retrouvé sur les parcours des éléphants peut s’expliquer par l’influence des actions des mega-herbivores tels que les éléphants à travers leur fort piétinement et forte pression sur le pâturage, ainsi que celles des feux d’aménagement. Les formations végétales exploitées par les éléphants dans la RBP sont dominées par les espèces soudaniennes et soudano-zambéziennes en termes de recouvrement. Ceci peut s’expliquer aisément parce que la RBP est située dans une zone soudanienne. Ce résultat est conforme aux résultats de Tchamba (1996) en zone savanicole où les espèces appétées par les éléphants sont en grande partie de type soudanienne ou soudano-zambéziennes.

De façon générale, ces résultats au niveau de la structure horizontale sont conformes à ceux de Gaoué (2000) et de Yabi (2002) qui trouvent que cette structure est normale car nous sommes en formation naturelle où la densité des jeunes arbres est généralement élevée. Cette décroissance rend compte de la forte prépondérance de la régénération dans les groupements étudiés. Si la structure est normale de façon générale, nous pouvons donc dire que les éléphants n’influencent pas trop la structure horizontale des peuplements. Ce constat rejoint celui de Pfeiffer (1989) qui trouve que l’éléphant fait preuve d’une souplesse écologique stupéfiante. Ainsi, au niveau du groupe C constitué par le groupement à Aneilema lanceolatum et Acacia goumaensis, l’effectif des arbres de diamètre supérieur à 30cm est considérable. Ce résultat rejoint ceux obtenus par Tchamba (1996) et Tchamba et Seme (1999) qui ont trouvé dans la région de Waza Logon, un milieu savanicole, que les éléphants sélectionnent certains grands arbres et les épargnent du fait de leur caractère toxique ou de la disponibilité alimentaire dont fait preuve le parc à l’endroit des animaux. C’est ainsi que plusieurs espèces sont épargnées. La faible représentativité des espèces à large distribution dont la prédominance dans une phytocénose est synonyme de dégradation (Sinsin 1993) permet de conclure que la flore exploitée par les éléphants dans la RBP conserve encore sa spécificité. Dans la zone à faible concentration, le phénomène contraire s’observe pour illustrer la faible pression due peut-être à la faible densité dans la zone et à la diminution des facteurs favorisant la présence des éléphants. Ces résultats s’expliquent aussi par les conclusions d’O’Coonor et al. (2007) qui précisent que les facteurs influençant la vulnérabilité des arbres à être mutilés incluent la force de l’arbre, la profondeur et l’extension du système


Aristide et al.

racinaire, de même que la stabilité du substrat.Dans la structure verticale, le groupe 1 est

caractérisé par l’inexistence des individus de petite taille (1 à 4m). L’éléphant dans cette zone de forte concentration casse l’arbre qui perd en hauteur. Ces résultats sont conformes avec ceux de Maire (2000) qui ont montré que l’éléphant exploite les ligneux entre 0 et 6m de hauteur. Ce phénomène observé pour les individus des classes comprises entre 10 et 12m peut être dû au fait que les éléphants par leur prélèvement sur les individus de basse classe empêchent leur croissance en hauteur. Concernant la structure verticale, dans les deux zones de concentration des éléphants, l’effectif des arbres de classe comprise entre 10 et 12m de hauteur étant faible dans la zone de forte concentration des éléphants, ceci dénote de la pression (chablis) enregistrée par les éléphants sur les espèces de cette classe. Cet état de choses n’étant pas observé dans la zone de faible concentration des éléphants, alors ces chablis peuvent être causés lors du passage de l’éléphant ou au moment du broutage pour l’alimentation de ses éléphanteaux. De plus, ces zones sont riches en espèces du genre acacia, alors que selon Kruger et al. (2007), les acacias se régénèrent malgré la coupure faite par les éléphants tandis que les autres espèces ne peuvent se régénérer. Aussi, dans la zone de forte concentration, la densité de ligneux est élevée et le taux de dégâts est élevé car, selon Law et al. (1975), Coe et al. (1976) et Cumming et al. (1997), en cas de fortes densités, les éléphants déciment généralement les terrains boisés qui deviennent des prairies plus clairsemées.

Conclusion

L’étude de l’identification et de la caractérisation des formations végétales exploitées par les éléphants dans la Réserve de Biosphère de la Pendjari permet de voir que les éléphants affectionnent plus les formations ouvertes que les formations fermées. Dans la Réserve de Biosphère de la Pendjari, ces formations sont les savanes arbustives, arborées localisées souvent dans des zones riches en espèces appétées et en mares temporaires et/ou permanentes. De même, les éléphants exploitent souvent des formations qui sont proches des zones anthropisées où se mènent des activités champêtres. Les analyses du point de vue floristique montrent que les éléphants de la RBP aiment et/ou préfèrent les zones ayant une diversité floristique importante. L`étude phytosocologique

permet de constater que les types de groupement observés et identifiés regorgent d`une diversité en espèces appétées par les éléphants.

Les analyses sur le plan dendrométrique montrent que les éléphants à la recherche d`une bonne alimentation qualitativement et quantitativement, et surtout sur les couloirs de passage détruisent les arbres qui perdent en hauteur (chablis) ou en diamètre (mutilation) dans la Réserve de Biosphère de la Pendjari.

Références bibliographiques

Alexandre, O.Y. 1978. Le rôle disséminateur des éléphants en forêt de Tai, Côte d’Ivoire. La Terre et la vie 32 : 47–72.

Alfa Gambari S. 2002. Suivi écologique des troupeaux d’éléphants d’Alfakorara dans la zone cynétrique de la Djona au Bénin. Mémoire de DIT. CPU/UAC. 99 p.

Alfa Gambari S. 2003. Déterminants du pouvoir germinatif des semences excrétées par les éléphants et la persistance des phytocénoses à l’irradiation par zoochorie : cas des éléphants et des phytocénoses du Parc Régional du W/Bénin et de ses zones périphériques. Mémoire de DEA. FLASH/UAC, 93 p.

Avohou, T.H. 2002. Détermination des potentialités pastorales de la chaine de l’Atacora au Nord–Ouest du Bénin, Région de TANGUIETA-BATIA. Thèse d’ing. agro., FSA/UAC, Bénin. 136 p.

Bonou, W., Glèlè, K.R, Assogbadjo, A.E, Fonton, H.N and Sinsin, B. 2009. Characterisation of Afzelia africana S.M. habitat in the Lama Forest Reserve of Benin. Forest Ecology and Management 258:1084–1092.

Braun-Blanquet, J. 1932. Plant sociology. The study of plant communities. Ed. New-York, London: McGray Hill. 439 p.

Coe, M. J., Cumming, D. H. and Phillipson, J. 1976. Biomass and production of large African herbivores in relation to rainfall and primary production. Oecologia (Berl.) 22:341–354.

Cumming, D.H.M., Fenton, M.B., Rautenback, I.L., Taylor, R.D., Cumming, G.S., Cumming, M.S., Dunlop, J.M., Ford, A.G., Hovorka, M.D., Johnson, D.S., Kalcounis, M., Mahlangu, Z. and Portfors, C.V.R. 1997. Elephants, woodlands and biodiversity in southern Africa. South African Journal of Science 93:231–236.

Gaoué, O.G. 2000. Facteurs déterminants pour le zonage de la zone cynégétique de la Pendjari comme base de



gestion intégrée. Th. Ing. Agr., FSA/UNB, Abomey-Calavi, Bénin. 106 p.

Houinato, M. 2001. Phytosociologie, écologie, production et capacité de charge des formations végétales pâturées dans la région des Monts Kouffé (Bénin). Thèse Doctorat. Fac. Se. Lab. Bot. Syst et phyt. Uni. Lib. Bruxelles, Belgique. 219 p.

Houinato, M., Sinsin, B. et Lejoli, J. 2001. Impact des feux de brousse sur la dynamique des communautés végétales dans la forêt de Bassila, (Bénin). Acta Botanica gacillia-cat.inist.fr .

Johnson, N.L. and Kotz, S. 1970. Distributions in statistics: Continuous univariate distributions. 2. New York: Wiley.

Kidjo, F.C. 1992. Ecodéveloppement rural d’Alfakoara (Djona). Eco-éthologie des éléphants (Loxodonta africana). SEAPA/PGRN/DFRN. 28 p.

Kruger, L.M., Coetzee, J.A. and Vichers, K. 2007. The impacts of elephants on woodlands and associated biodiversity. Summary report to South African National Parks. Organization of Tropical Studies.

Laws, R.M., Parker, I.S.C. and Johnstone, R.C.B. 1975. Elephants and their habitats. The ecology of elephants of north Bunyoro, Uganda. Oxford, UK: Clarendon Press. 376 p.

Legba, F. 2005 Contribution de la végétation des collines de la zone cynégétique et du Parc national de la Pendjari du Bénin comme milieu ressource de la faune sauvage. Th. Ing. Agr. FSA/UAC/Bénin. 121 p.

Maire, M. 2000. Impact actuel des éléphants sur la savane à Acacia seyal : Parc national de Zakouma (Sud-est du Tchad), N’Djaména, Tchad. Fif-Engref/Commission Européenne. 34 p.

Mubalama, K.L. 2000. Les relations hommes-éléphants dans la Réserve de Faune à Okapis (Okapia johnstonii) en République démocratique du Congo. Nature et faune 16(2) : 19–34.

Natta, A.K. 2003. Ecological assessment of riparian forests in Benin: phytodiversity, phytosociology, and spatial distribution of tree species. PhD thesis, Wageningen. 205 p.

O’Connor, T.G. 2007. A functional hypothesis of the threat of local extirpation of woody plant species by elephant in Africa. Biological Conservation 136: 329–345.

Pfeiffer, P. 1989. Vie et mort d’un géant. L’éléphant d’Afrique. L’odyssée, Flammarion. 192 p.

Pielou, E.C. 1966. Species diversity and pattern diversity in study of ecological succession. Theor. Biol. 10 : 370–383.

Poole, J. 1996. L’éléphant d’Afrique. Nairobi, Kenya: African Wildlife Foundation. pp. 1–7.

Sinsin, B. 1993. Phytosociologie, écologie, valeur pastorale, productivité et capacité de charge des pâturages naturels du périmètre de Nikki-Kalalé au Nord Bénin. Thèse doctorat. Université Ub. Buxelles, Belgique. 390 p.

Sokpon, N., Biaou, H., Gaoue, O.G., Hunhyet, O.K., Ouinsavi, C. et Barbier, N. 2001. Inventaire et caractérisation des formations végétales du Parc national de la Pendjari, Zones cynégétiques de la Pendjari et de l’Atacora (Région de Konkombri). Unité de Sylviculture et d`Ecologie Forestière/FSA/UNB/République du Benin. 48 p.+annexes.

Shannon, C.E., and Weiner, W. 1963. Mathematical theory of communication. Illinois, USA: University of Illinois Press, Urbana.

Sinsin, B., Akpona, H. et Ahokpe, E. 2006. Dénombre-ment aérien de la faune dans la Réserve de Biosphère de la Pendjari. Rapport technique. CENAGREF/ Projet Pendjari – GTZ – GFA Consulting. Cotonou, Benin. 35 p.

Tchamba, M.N. 1996. Elephants and their interaction with people and vegetation in the Waza-Logone Region, Cameroon. PhD thesis, University of Utrecht, Netherlands.

Tchamba, M.N. and Seme, P.M. 1999. Diet and feeding behavior of the forest elephant in the Santchou Reserve, Cameroon. African Journal of Ecology 31:165–171.

Tehou, A.C. 1995. Etude écologique des éléphants (Loxodonta africana) d’Alfakoara. PGRN/VGEFER. 41 p.

Tehou, A.C. 2001. Mode de dissémination des espèces les plus appétées par les éléphants dans la zone cynégétique de la Djona, les forêts classées de Goungoun, de la Sota et des environs, Nord-Bénin. Pachyderm 30 : 65–69.

Tehou, A.C. 2002. Les éléphants (Loxodonta africana) dans la Reserve de Biosphère de la Pendjari, Nord-Ouest République du Bénin : abondance, densité et répartitions spatiales. Pachyderm 33 : 64–68.

Tehou, A.C. et Sinsin, B. 1999. Elephants and their interactions with people and vegetation in the Waza-Logon region. Cameroun, NUFFIC. 230 p.

Tehou, A.C. et Sinsin, B. 2000. Ecologie de la population d’éléphants (Loxodonta africana) de la zone cynégétique de la Djona (Bénin). Mammalia 64 (1) : 29–40.


Aristide et al.

Wala, K. 2004. La végétation de la chaîne de l’Atacora au Bénin : diversité floristique, phytosociologie et impact humain. Thèse doctorat. Université de Lomé. Togo : 138 p + annexes.

White, L.J.T., Titin, C.E.G. and Fernanders, M. 1993. Group composition and elephant diet on forest

elephants, Loxodonta cyclotis Matshie 1900, in the Lope Reserve, Gabon. African Journal of Ecology 31:181–199.

Yedomonhan, A.P. 2002. Etude de la végétation sociale du secteur méridional du centre Bénin. Mémoire FLASH/UAC, Bénin. 94 p.


Optimizing the habitat of the Javan rhino in Ujung Kulon National Park

Optimizing the habitat of the Javan rhinoceros (Rhinoceros sondaicus) in Ujung Kulon National Park by reducing the invasive palm Arenga obtusifolia

Adhi R.S. Hariyadi,1 Agus Priambudi,2 Ridwan Setiawan,1 Daryan,2 Hendra Purnama 2 and Asep Yayus 2

1 WWF Indonesia – Ujung Kulon Project, Jl. Perintis Kemerdekaan no 104, Labuan–Pandeglang, Banten–42264, Indonesia2 Balai Taman Nasional Ujung Kulon, Jl. Perintis Kemerdekaan, Labuan–Pandeglang, Banten–42264, Indonesia

AbstractThe dominance of the arenga palm had been identified as a potential limiting factor in the distribution of Javan rhinos because it prevented the growth of food plant species, thus created a nutrition-poor area in which rhinos roamed. Palm dominance was reduced in an area consisting of four 1-ha plots (assigned as plots A–D) that was selected on the basis of various ecological criteria such as number of palm trees (shading) and potential for connecting home ranges of several rhinos. The research was carried in two phases—palm control and monitoring—to study the impact of palm control on the growth of food plants and, consequently, on the area’s accessibility by rhinos. Palm control resulted in increasing abundance and diversity of food plants for rhinos, and increasing visitation by rhinos. Results from this research show that controlling palm and other invasive plant species can potentially be used to improve the Javan rhinoceros’s accessibility to certain areas within the habitat.

RésuméLa domination du palmier Arenga avait été identifié come un facteur limitatif potentiel dans la distribution des rhinocéros de Java, car il empêche la croissance des espèces de plantes alimentaires et crée une zone « pauvre en nurition » pour les rhinocéros. La réduction de la domination du palmier s’est effectuée dans une zone composée de quatre parcelles d’un ha sélectionées sur base de divers critères écologiques tels que le nombre de palmiers (ombrage) et le potentiel de relier les habitats vitaux de plusieurs rhinocéros. La recherche a été effectuée en deux phases—le contrôle de palmier Arenga a entraîné une augmentation de l’abondance et de la diversité des plantes alimentaires pour les rhinocéros ainsi qu’une augmentaiton de la fréquentation par les rhinocéros. Le résultat de cette recherche montre que le contrôle du palmier et des autres espèces de plantes envahissantes pourrait être utilisé pour améliorer l’accessibilité des rhinocéros à certaines zones au sein de leur habitat.

IntroductionAfter enduring a period of a very small population with only 25 animals in 1967, the Javan rhinoceros (Rhinoceros sondaicus) reached a population of 58 in 1980, but unfortunately there has not been a significant increase in population size since then (Setiawan et al. 2002; Groves & Leslie 2011). Many studies have been conducted in Ujung Kulon National Park (NP) to determine the limiting factors causing the population to stagnate. The factor suspected to be the major setback is habitat conditions that reduce

the park’s carrying capacity. The carrying capacity of habitats for herbivores such as the Javan rhinoceros is determined by many factors, such as rainfall and soil quality, which eventually influence the abundance and diversity of food plants (Fritz & Duncan 1994) as well as their nutritional quality (Hobbs & Swift 1985). Low food quality is a threat that should be avoided when managing wild herbivore populations as it may result in mortality during the first six months of an animal’s life due to lack of nutrition in the food supply (Freeland & Choquenot 1990). It is also important to note that in the context of large herbivores, access


Hariyadi et al.

to food sources such as feeding ground plays an important role in determining the limit of the habitat’s carrying capacity (Boone & Hobbs 2004). Previous studies done by Bogor Agricultural University indicate that the dominance of arenga palm, Arenga obtusifolia, a rapidly spreading palm species, poses a threat by reducing the availability of food plants for rhinos. Therefore, in addressing the issue of availability of food plants, it is important to reduce the dominance of this invasive palm in certain areas within Ujung Kulon peninsula to allow the growth of other plant species with high palatability for Javan rhinos (Putro 1997).

Arenga palm is classified under the Arecaceae (or Palmae) family (Sastrapradja 1978) that relies on both root extension and seed for spread and dispersal. The arenga palm was not recorded in Ujung Kulon NP in the 1980s (Hommel 1987), but it is now found in most places within the park. Ripe fruits are normally consumed by palm civets (Paradoxurus hermaphroditus) and the long-tailed macaque (Macaca fascicularis) that help in seed dispersal by expelling palm seeds through faeces or by dropping seeds after consuming the fruit (Lucas & Corlett 1992; Putro 1997). Besides seed dispersal, the arenga palm can also spread and sprout using its extensive subterraneous roots that also restrict the growth of other plant species (Supriatin 2000). These dispersal methods make the arenga palm superior over other species in the course of vegetation succession in Ujung Kulon. As its mode of expansion and dispersal is known, it is important to ensure that the palm’s fruits and roots are removed completely or destroyed to minimize the probability of it regrowing. This can be done by removing it manually or by using chemicals such as isopropyl ammonium glyphosate, an active ingredient of common weed killer, that has no negative effects on the natural flora (Carlson & Gorchov 2004) even after a prolonged five-year use (Hochstedler et al. 2007). Unfortunately, despite the potential of isopropyl ammonium glyphosate to effectively and efficiently eradicate the invasive arenga palm, the relevant authorities have note approved its use inside Ujung Kulon NP and all efforts to control the arenga palm must resort to manual interventions.

Materials and methodsAn area measuring 4 ha was selected in the peninsula of Ujung Kulon NP (6º43ʹ14.8ʺ–6º43ʹ17.3ʺ S and 105º19ʹ12.3ʺ–105º19ʹ12.3ʺ E) based on criteria that

included known presence of rhinos in the vicinity of the plot based on previous observation and census data; presence of other determinants for rhinos such as wallow holes, feeding ground, etc.; high palm density of more than 700 palm trees/ha (causes shading that prevents the growth of other plant species); and potential viability as a corridor to connect rhino home ranges in the peninsula of Ujung Kulon NP. The habitat was manipulated for 17 months from February 2008 until June 2009. The work schedule consisted of: phase 1, clearing between February and April 2008, phase 2, treating in May–November 2008, and phase 3, monitoring from December 2008 to June 2009. Specific activities included systematic clearing of a palm-infested plot, monitoring the growth of food plants important to the Javan rhinoceros, and monitoring rhino visitation to the treated plot. The 4-ha area designated as the study area was divided into four plots each measuring 1 ha. Each of the four plots (plots A to D) was further divided into smaller subplots of 20 m x 20 m containing smaller grids—2 m x 2 m, 5 m x 5 m, and 10 m x 10 m—to allow vegetation analysis of four classes of growth stages: sapling, hedge, young tree, and full-grown tree. Each subplot was used to calculate the density of full-grown tree densities, the 10 m x 10 m grids were used to calculate density of the young trees, the 5 m x 5 m grids were used for shrubs, and the 2 m x 2 m grids were used to calculate sapling densities in each subplot (Fig. 1). Palms were cleared within the first five months of the study by a team of 10 persons: two persons each operating a Motoyama 24-inch chainsaw, and eight persons removing the fallen palms out of the optimization plot to prevent them regrowing through seed dispersal. Next, palm roots and stubs were removed from the plot using shovels and hatchets to prevent arenga palm regrowing from roots. The effect after treatment on food plants important to rhinos was reflected in the diversity of food plants (numbers of plant genera) and abundance (quantity of food plants available).

To study the nutritional value of the new food plants growing on the plot, proximate analyses were conducted with gravimetric, acid-base extraction, bomb calorimeter, auto analysis destruction, Atomic Absorbance Spectrophotometry, and conventional spectrophotometry, specifically water, fat, ash, rough fibres, energy, protein content, calcium, and phosphate-tannin. All food plant samples were sun dried for 8 hours and oven dried at 60 ºC until they reached constant weight before the proximate analysis.



An average of three random observation points was set up monthly within the 4-ha study area to observe rhino presence from 0800 h to 1730 h in the period between 19 August 2008 and 17 July 2009. The frequency of rhino visits to the 4-ha plot was calculated as the number of observation periods with rhino presence divided by total number of observation periods of rhino visits in the treated plots. This information was compiled to analyse how rhinos used the treated areas every month during the monitoring phase of this study. Rhinos were identified by the footprint size and visually using a DVREye automatic video camera (Pix Controller, Inc., 1056 Corporate Lane Murry Corporate Park Export, PA. USA), while plots, subplots and rhino presence were marked using a Garmin 12 XL global positioning device (GPS) to obtain all coordinates. Data from vegetation analyses and rhino visitations were compiled using spreadsheet software Microsoft Excel (Microsoft Inc.), and statistical analyses were done using biostat software

(Analystsoft). One-way analysis of variance (ANOVA) was used to test the null hypotheses that there would be no significant differences in the numbers and diversity of sapling presence before and after treatment.

ResultsSystematic clearing of the area dominated by the arenga palm resulted in an increase in the number of saplings (Fig. 2). ANOVA rejected the null hypothesis and indicated significant differences between the number of saplings before clearing between February and April and after clearing phases in May–November (df = 1; Fcrit = 5.9874; F = 21.8474), but not between clearing in May–November and monitoring phases in December–June (df = 1; Fcrit = 5.9874; F = 4.2505). ANOVA also rejected the null hypothesis

and suggested that the diversity of newly growing saplings differs significantly between February–April and May–November phases (df = 1; Fcrit = 5.9874; F = 93.0258) as well as between May–November and December–June phases (df = 1; Fcrit = 5.9874; F = 14.2274). Descriptive analysis showed that, unlike the numbers of saplings growing in the plot area, the diversity of saplings (indicated by the numbers of different plant species) continued to increase significantly throughout these periods. In the initial period, vegetation analysis showed an average of 42.5 plant species per plot, while the later period showed an average of 65 plant species per plot. Table 1 gives the plant species found in the palm control plots, Table 2 the nutritional values of some of the rhino food plant species growing in this area. Rhino visitation data showed a trend of increasing occurrence throughout the year (Fig. 2), and at least two different rhinos were identified through the automatic video camera installed in the combined 4-ha study area. Frequency

Figure 1. The plot layout for vegetation measurements on the 4-ha study area, an area of 1 ha (100 m x 100 m) divided into smaller grids of 2 m x 2 m (A), 5 m x 5 m (B), 10 m x 10 m (C), and 20 m x 20 m (D) to properly assess various stages of vegetation consisting of seedling, hedge, young tree, and full-grown tree.


Hariyadi et al.

occasion) in the monitoring phase between December 2008 and February 2009 and in April 2009.

Discussion

Statistical analysis confirmed that the number of arenga palms in all four plots (plots A–D) was significantly reduced and was followed by the growth of non-palm saplings that included rhino food plant species. The pattern of sapling growth in the palm control plot showed similarity with a succession where a period of growth was followed by a period of growth stagnancy, indicating that the optimum capacity of growth was reached. Furthermore, the composition of plant species found in the area showed a trend of increasing diversity (increasing numbers of species) throughout the observation periods. These facts suggest a direct correlation between the decreasing palm populations with the increasing growth of other plant species. The plant species were dominated by Spondias pinnata, a relatively good source of water and calcium for the rhinos but a poor source of energy (lower calorie content than that of other food plant species). The plants showed lower water content than they normally would in natural state due to the drying processes; therefore the water content from proximate analysis was suitable for indicating the water content that could be retained by these plant species.

Increased growth of plants known to be food sources for Javan rhinos seemed to influence the rhinos’ use of the treated plots. This can be inferred from the data showing the increase in rhino visitation (percentage of rhino occurrences in the area) at

of rhino visitation indicated no visitation in August and September 2008 (treated phase), and showed steady visitation by two individuals (footprint size 25–26 cm and 27–28 cm) in treated and monitoring phases between November 2008 and July 2009 with most occurrences (100% rhino findings in every observation

Table 1. Plant species known to be food plants for Javan rhinos found growing as seedlings in the cleared area.

Food plant species No. of stems found in plot

Spondias pinnataLagerstroemia speciosaMikania cordataMusa sp.Unclassified 1Mallotus floribundusUnclassified 2Cordia dichotomaZanthoxylum rhetsaEupatorium odoratumLantana camaraUnclassified 3Tetrastigma lanceolariumSolanum sp.Unclassified 4Donax canniformisGlochidion rubrumAmomum sp.Unclassified 5Trema orientalisSolanum torvumAnthocephalus chinensisDillenia obovataPterocymbium tinctoriumLepidagathis javanica

1152417151198877654332222211111

A. Feb-Apr, 16 A. May-Nov, 283

A. Dec-June, 713B. Feb-Apr, 31 B. May-Nov, 198

B. Dec-June, 416C. Feb-Apr, 28

C. May-Nov, 305

C. Dec-June, 506

D. Feb-Apr, 23 D. May-Nov, 498

D. Dec-June, 423

A. May-Nov, 38A. Dec-June, 75

B. May-Nov, 35

B. Dec-June, 58C. May-Nov, 42

C. Dec-June, 68

D. May-Nov, 55

D. Dec-June, 59

A B

Figure 2. Bar graph indicates the numbers of seedlings (A) and the diversity of seedling species (B) growing in the areas before palm clearing (February–April), after clearing (May–November), and at maintenance period (December–June). The data show information from all four plots (plots A–D) combined; no data were collected for seedling diversity during the February–April period.



approximately the same time as the increase in growth of food plants. Rhinos were first identified by their footprints; the footprint sizes suggested two animals were entering this study area. Further observation using video trap revealed that the two rhinos were a mother travelling with her calf (Fig. 4). Results from field observations were supported by video clip records of the rhinos showing them using these plots as a new path as well as new feeding ground. Since no rhino presence had been recorded previously in this particular area, it can be concluded that the palm control area provided a new access for certain rhinos. Therefore, habitat optimization by reducing the amount of invasive palm (or any other plant) species could be used by park managers as a tool to increase access and probability of joining the isolated rhinos for breeding purposes and for improving the quality of rhino home ranges.

In addition to manually removing the arenga palm, there are options for using

chemical as well as biological agents to control its spread. The use of biological agents (insects) has been implemented and evaluated for controlling Lantana camara in Australia (Day et al. 2003), and a similar concept may be applied to control the arenga palm. Potentially invasive plant species that could affect the

Figure 3. Occurrence of rhino visitation (%) in three observation points in all cleared plots (combined) from August 2008 to July 2009.

Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Jun. Jul.

Plo

ts v

isite

d b

y rh

inos

(vis

itatio

n fr

eque

ncy)

%

Observation months

50% 50% 50% 50%

40%

100%100%100%

0% 0% 0% 0%

20

40

60

80

100

0

Table 2. Proximate analysis of various plant species

Plant speciesCalorie kcal/kg

Grams per 100 grams

Water Protein FatRough fibre Ash C P Tannin

Spondias pinnata 3.005 9.42 9.16 2.62 15.57 20.16 4.70 0.19 0.02Mallotus floribundus 3.521 7.89 15.71 4.45 25.93 11.42 2.42 0.30 0.91Amomum sp. 4.151 9.58 10.24 1.63 33.79 8.69 0.67 0.23 0.18Lantana camara 4.004 8.37 7.67 4.11 15.96 9.99 2.04 0.16 0.05

Figure 4. Rhino mother and calf using the study area as the new feeding ground.

© U

jung

Kul

on N

P Au

thor

ity/W

WF-

Indo

nesi

a


Hariyadi et al.

habitat quality of the Indonesian rhinoceros included vine (Merremia peltata) and wild bamboo species such as Schizostachyum zollingeri and Bambusa sp., as well as Lantana camara.

AcknowledgementsThis research was supported by the Ujung Kulon National Park Authority and WWF-Indonesia with funding from WWF-Germany and WWF-UK. Video trap equipment was provided by the International Rhino Foundation and the Asian Rhino Project.

ReferencesBoone, R.B. and Hobbs, N.T. 2004. Lines around

fragments: effects of fencing on large herbivores. African Journal of Range and Forage Science 21(3):147–158.

Carlson, A.M. and Gorchov, D.L. 2004. Effects of herbicide on the invasive biennial Alliaria petiolata (garlic mustard) and initial responses of native plants in a south-western Ohio Forest. Restoration Ecology 12(4):559–567.

Freeland, W.J. and Choquenot, D. 1990. Determinants of herbivore carrying capacity: plants, nutrients. Equus asinus in northern Australia. Ecology 71(2):589–597.

Fritz, H. and Duncan, P. 1994. On the carrying capacity of the large ungulates of the African savannah ecosystems. Proceedings of the Royal Society, London B. 256:77–82.

Day, M.D., Broughton, S. and Hannah-Jones, M.A. 2003. Current distribution and status of Lantana camara and its biological control in Australia with recommendations for further biocontrol introductions into other countries. Biocontrol News and Information 63N–76N.

Groves, C.P. and Leslie, D.M. 2011. Rhinoceros sondaicus (Perissodactyla: Rhinoceroidae). Mammalian Species 43(887):190–208.

Hochstedler W.W., Slaughter B.S., Gorchov, D.L., Saunder, L.P. and Stevens, M.H.H. 2007. Forest floor plant community response to experimental control of the invasive biennial Alliaria petiolata (garlic mustard). Journal of the Torrey Botanical Society 134(2):155–165.

Hobbs, N.T. and Swift, D. 1985. Estimates of habitat carrying capacity incorporating explicit nutritional constraints. Journal of Wildlife Management 49(3):814–822.

Hommel, P.W.F.N. 1987. Landscape ecology of Ujung Kulon (West Java, Indonesia). PhD dissertation. Soil Survey Institute, Wageningen.

Lucas, P.W. and Corlett, R.T. 1992. Notes on the treatment of palm fruits by long-tailed macaque (Macaca fascicularis). Principes 36(1):45–48.

Putro, H.R. 1997. Heterogenitas habitat badak Jawa (Rhinoceros sondaicus Desm. 1822) di Taman Nasional Ujung Kulon. Media Konservasi edisi khusus (1997):17–40.

Sastrapradja, S. 1978. Palem Indonesia. Lembaga Biologi Nasional – LIPI, Bogor.

Setiawan, R., Yahya, A., Hariyadi, A.R.S. and Polet, G. 2002. Population and distribution of Javan rhinoceros (Rhinoceros sondaicus, Desmarest, 1822) based on the calculation of fecal collection in Ujung Kulon National Park. WWF Report. Cooperation WWF Indonesia and Ujung Kulon NP Authority Project ID 0091.07.

Supriatin. 2000. Studi kemungkinan adanya pengaruh alelopati langkap (Arenga obtusifolia Blumme ex. Mart) terhadap pertumbuhan semai tumbuhan pakan badak Jawa (Rhinoceros sondaicus Desmarest 1822) di Taman Nasional Ujung Kulon. PhD dissertation. Institut Pertanian Bogor, Bogor, Indonesia.

Demand for forest elephant ivory in Japan



Tomoaki Nishihara

Senior Technical Adviser of Operations and Protection, Wildlife Conservation SocietyBP 14537, Brazzaville, Republic of Congoemail: [email protected]

AbstractJapan is the only country where a strong demand for ‘hard’ or forest elephant ivory still exists. This demand differs from that of China, which consumes more ivory but where no preference for soft or hard ivory exists. While there is no hard evidence that ivory originating from forest elephants in Central Africa is smuggled into Japan, the question remains whether only old stock of hard ivory can be meeting the stable demand. Through investigations in ivory markets in Japan, it was found that ivory dealers there have limited knowledge of the domestic ivory-trade control system and have not applied it. Also, this control system is not sufficient to manage the legal ivory trade. These factors may facilitate the illegal importation of hard ivory into the Japanese market. We recommend that the Japanese ivory-management system be re-evaluated and improved, focusing on hard ivory stock management. In addition, Japan should develop an information-sharing system for forest elephant conservation by producing and distributing practical education materials in Japanese. These are priorities because hard ivory originates in the Central African region, where poaching pressure on forest elephants is increasing, resulting in a drastic decline in their populations.

RésuméLe Japon est le seul pays où il existe toujours une forte demande pour l’ivoire «dur» ou l’ivoire des éléphants de forêt. Cette tendance diffère de celle de la Chine qui a une plus grande consommation d’ivoire, mais où il n’existe aucune préférence pour l’ivoire souple ou dur. Bien qu’il n’y ait pas de preuves tangibles que l’ivoire provenant des éléphants de forêt en Afrique centrale passe en contrebande au Japon, il reste la question de savoir si seulement le vieux stock d’ivoire dur pourrait répondre à la demande stable. Grâce à des enquêtes sur les marchés d’ivoire au Japon, on a constaté que les négociants d’ivoire au Japon ont une connaissance limitée du système de contrôle interne du commerce de l’ivoire et ne l’appliquent pas. En outre, ce système de contrôle ne suffit pas à gérer le commerce légal de l’ivoire. Ces facteurs peuvent faciliter l’importation illégale d’ivoire dur sur le marché japonais. En conclusion, le système de gestion de l’ivoire japonais doit être réévalué et amélioré, et l’accent mis sur la gestion des stocks d’ivoire dur. En outre, le Japon devrait mettre au point un système d’échange d’informations sur la conservation de l’éléphant de forêt, en produisant et en distribuant du matériel didactique pratique aux Japonais. Tout cela est prioritaire parce que l’ivoire dur provient d’Afrique centrale, où la pression du braconnage sur les éléphants de forêt s’accroît ce qui donne lieu à une baisse drastique de leurs populations.

IntroductionTRAFFIC—the wildlife-trade monitoring system—states that 2011 saw the largest seizures of ivory by weight since the CITES (Convention on International Trade in Endangered Species of Wild Fauna and Flora) ban on ivory trade in 1989 (TRAFFIC 2011). While most of this ivory originated in Africa there is no indication of its precise origins. These seizures are

related to the current high demand for ivory in Asia, particularly from China (Martin & Vigne 2011; Gabriel et al. 2012). It is highly probable that part of the seized ivory came from forest elephants (Loxodonta africana cyclotis) if we consider their recent population decline (Beyers et al. 2011; Bouche et al. 2011; Maisels et al. in review) and high poaching pressure for their ivory.

Historically, Japanese ivory carvers have produced most of their items from ‘hard ivory’ originating from

Nishihara


Asian elephants and from forest elephants of Central Africa (Martin 1985), which is preferred to ‘soft ivory’ from savannah elephants (Loxodonta africana africana) of eastern, western and southern Africa. The major ivory items produced in Japan are hanko (personal name seals) and musical implements, such as bachi (plectrum for the shamisen, a traditional Japanese musical instrument), preferably made of hard ivory (Vigne and Martin 2010).

However, due to the CITES ivory ban of 1989, hard ivory could not be imported to Japan, resulting in a drop in amount of its stock. With two one-off ivory sales under CITES regulations, in 1999 and in 2008, Japan acquired a quantity of soft ivory. Since the ban, more and more ivory items in Japan were therefore made from soft ivory, but most Japanese ivory dealers consider soft ivory useless for traditional carving, particularly for bachi (Nishihara 2003). Although Japanese consumer demand for hanko made of ivory has been declining since the last decade (Vigne and Martin 2010), it is clear that there is still continuous demand for ivory (Motegi 1988; Tsugawa 2004; Tanaka 2008), especially for hard ivory (Tanabe 1963) from the Central African region for bachi (Association of Japanese Actors 2000; Tanaka et al. 2009).

The major objective of this study was to examine the current status of demand for hard ivory in Japan, and to investigate knowledge about elephants and ivory as well as about the Japanese domestic management system of ivory by ivory dealers in Japan. At the same time, the Japanese ivory management system was examined, with suggestions for improving its functioning to control illegal importation and use of ivory more efficiently. This study is based mainly on descriptive information because the author was unable to collect quantitative data on hard ivory, its products and its demand due to a lack of information from Japanese authorities and bachi-related people.

Materials and methodsInvestigations in internet and ordinary shops in Japan that deal in ivory products, particularly hanko and bachi, were conducted by three Japanese investigators under supervision of the author, during two periods

in 2010—March–April and August–September—in Tokyo, Yamanashi Prefecture, where traditionally many ivory hanko have been produced, and in the neighbouring prefectures of Kanagawa, Saitama and Gunma. In this report, the shops selected were in this area of Japan; most internet shops are located in the western part of Japan.

The investigators posed as clients, aware that most of the shopkeepers were wary about discussing ivory issues. This sensitive reaction may be related to poor understanding of the legal sale of ivory products in Japan and the international ban on ivory trade. Table 1 shows the number of shops investigated. Also, during the period November 2009–October 2011, any information related to Japanese ivory issues was collected from books, journals, newspapers and websites.

In ordinary hanko shops, the investigators asked three categories of questions:1. Current trends in hanko trade• Prices of hanko made of any material including

ivory• Trends in prices, demand and supply, age class of

buyers, seasonal variations and annual variations of ivory sales

2. Knowledge of elephants and ivory• Knowledge of the geographical origin of ivory• Knowledge of ivory in general, for instance, how

it is harvested, whether ivory is regenerated after it is removed from the elephant

• Knowledge of hard and soft ivory3. Knowledge of the legal issues of ivory• CITES and Japanese domestic management system

of ivory• Presence of CITES certificate seal (Fig. 1)—the

official seal with the CITES logo provided by the relevant Japanese ministries

One investigator carried out research on internet hanko shops by checking for any information including hanko prices.

Table 1. Shops and other sources interviewed during this investigationPeriod in 2010 Ordinary

hanko shopsInternet hanko shops

Shamisen shops

JWRC and TIACA

March and April 66 23 12 2

August and September 36 10 17 0

Total 102 33 29 2

JWRC – Japan Wildlife Research Center; TIACA – Tokyo Ivory Arts and Crafts Association



For shamisen shops, one investigator posed the same questions used in hanko shops. Apart from ivory shops, one investigator visited ivory-related organizations to gather information on ivory management and legal issues. These organizations are 1) the Japan Wildlife Research Center (JWRC), the governmental organization that hosts the department of CITES, and manages and registers wildlife products under the Ministry of Environment and the Ministry of Economy, Trade and Industry, and 2) the Tokyo Ivory Arts and Crafts Association (TIACA), which has four

business activities—purchasing raw ivory, holding meetings to exchange the ivory and products, advertising ivory business, and holding exhibitions for immediate ivory sale (http://www.tokyo-ivory.or.jp/englishversion/index.html).

Results

Price of hanko and current trend in the hanko marketCurrently, the common material for hanko is horn of the Asian domestic water buffalo, Bubalis bubalis, because it is much cheaper than ivory but aesthetically pleasing and strong enough for use as hanko. Almost half of the ordinary hanko shops investigated (44 of 102 shops) explained this current trend. Titanium as hanko material is also recommended by some shops, but it is expensive and has yet to gain popularity among the Japanese. Table 2 compares maximum and minimum prices of Jitsu-in hanko (officially individually registered

seal) of typical 15-mm diameter made from various materials, indicating current price trends. The greater range exists between minimum and maximum prices for hanko made of ivory and buffalo horn, but this is because of the quality differences between parts of materials. Sample numbers of hanko made with titanium were few and therefore the prices cannot be considered indicative. The minimum price of ivory hanko seems to be cheap, especially in the internet shops. One ivory dealer said that it is impossible for

prices of ivory hanko in the internet shops to be so low if we see the ordinary wholesale prices, indicating that internet hanko shops use illegal ways to obtain ivory materials more cheaply because they are not well controlled.

Other current trends such as age class of buyers, seasonal variations and annual variations of ivory sales were not clearly detected during this investigation.

Table 2. Maximum and minimum prices of Jitsu-in hanko (officially individually registered seal) with typical 15-mm diameter, made from various materials as sold in ordinary and internet hanko shops

Material / type of shop Hanko investigated (no.)

Price (JPY)Minimum Maximum

Ivory Ordinary 86 11,800 110,350 Internet 93 9,800 273,750Horn of black buffalo Ordinary 12 9,980 46,220 Internet 41 1,680 30,450Titanium Ordinary 1 15,555 15,555 Internet 6 7,600 38,750

Approximately JPY 90 = USD 1 during the investigation period

Figure 1. CITES certificate seal attached to ivory products in Japan with the CITES logo (about 2.5 cm x 2.5 cm), with the wording ‘Certificate seal is proof that ivory products are legal based on the Law for the Conservation of Endangered Species of Wild Fauna and Flora (LCES) (from a brochure published by the Ministry of Environment and the Ministry of Economy, Trade and Industry, 2008).

Nishihara


Investigation of ordinary hanko shops about knowledge of elephants and ivoryIt is important to know the extent of awareness of ivory legal issues and also knowledge of elephants and ivory. This knowledge is essential for interactions between vendors and clients because with accurate information about elephants, ivory vendors can explain ivory products to clients with more transparency. It is indispensable for vendors to have this knowledge as now since the ivory ban most clients in Japan hesitate to buy any ivory products. Clients know little about elephants and ivory, and they request clear explanations from vendors. The following information was obtained from the shopkeepers:• Professional ivory carvers can traditionally distin-

guish between hard and soft ivory, but more than 70% of the shops cannot, which means they do not know the difference in properties or from which species each ivory type derives.

• About two-thirds of the shops did not know the geographical origin of elephant ivory; only one-third knew that it came from Africa but even they had no specific information about areas in Africa.

• Two shopkeepers believed that elephant tusks re-generate even after they are completely removed from the elephant. Only 11% of shops knew that the tusks do not regenerate.

• One shopkeeper believed that tusks are pulled out completely from living elephants without killing them.This information highlights the lack of knowledge

about ivory in Japan. Many of the shopkeepers are not aware of the existence of hard ivory, the origin of ivory, or the fate of elephants.

Investigation of shamisen and of shops selling Japanese musical instruments about knowledge of elephants and ivoryThe shamisen (Fig. 2) is one of the most popular traditional Japanese instruments, dating back more than 500 years, being used in Japanese performance arts such as Kabuki and Bugrake, both with 400 years of history (Tsugawa 2004; Tanaka et al. 2009). Bachi (Figs. 3 and 4) made from ivory started being used during the later age of the Edo period (Tanabe 1963) and became popular among professional shamisen musicians after the Meiji era, about 150 years ago. More than 60% (18 of 29) of shamisen shops

investigated mentioned this history of bachi.People insist that hard ivory is still available in

Japan from old stock acquired before the CITES ban. But the author was unable to procure information from the traders on hard ivory stocks in Japan including whether they contain large tusks. The average weight of ivory tusks, either hard or soft, in Japanese stock between 1995 and 2007 was only 12.86 kg (data from JWRC 2007); one bachi requires a large tusk weighing more than 15 kg (Fig. 5), ideally without any cracks. During the study period, the author found bachi made with tusk fragments because they could not make a whole bachi from cracked ivory, but the vendor explained that this was not ideal for professional users.

A bachi made of hard ivory provides better and softer sound; bachi made of soft ivory does not produce good sound, according to Japanese shamisen players from two different blogs. Bachi made of alternative materials such as synthetic resin have been used but

Figure 2. Shamisen, one of the most traditional and popular Japanese musical instruments, with bachi, a shamisen plectrum of ivory.

© P

IXTA



hanko shops interviewed. This is probably linked to the exclusive use of hard ivory for bachi. General knowledge of ivory—its geographical origins, elephant sub-species and habitat—is still poor.

Evaluation of the Japanese ivory management system

Japan established its own domestic trade management system of ivory to prevent illegal ivory imports into Japan, under the Law for the Conservation of Endangered Species of Wild Fauna and Flora (JWCS 2000, 2002; Martin & Stiles 2003). The system requires that all ivory and ivory products be strictly registered at every link along the commodity chain: importers, retailers, carvers and vendors. Nevertheless, loopholes still exist.

All dealers, at whatever level in the chain, must register each ivory item using a document with a unique number. This number should enable each item to be tracked, verifying that the product comes from legal ivory. However, no computer database exists for this system, only a paper numbering system with the dealers (JWCS 2000). Thus it is almost impossible to track the origin of any particular ivory product and extremely difficult to judge its legality.

All ivory dealers in Japan are required to be authorized by the Ministry of Environment and the Ministry of Economy, Trade and Industry. An authorized certificate with dealer’s number is given to each authorized ivory dealer (JWCS 2002). However, our investigations determined that the certificate papers were confirmed in fewer than half of the shops:

Figure 3. Ivory art technician, Kyoto, Japan, during the Edo period. The craftsman is cutting the ivory tusk into pieces; the shamisen bachi (plectrum) are at the lower right. Source: Jin Rin Kin Mou Zui (in Japanese)

Figure 4. Bachi, shamisen plectrum, made of wood; the size of bachi varies but the ordinary length is about 25 cm, maximum width 15 cm and maximum thickness 2.5 cm.

© T

amik

o Ta

mur

a

more research needs to be done, particularly with regard to the elasticity of the bachi edge. The future for traditional music in Japan, particularly music of high-quality sound, is not favourable due to the shortage of materials such as hard ivory (Tanaka et al. 2009).

A stable demand for bachi exists from professional shamisen musicians because they change bachi regularly. Musicians vary widely as to how often they change ivory bachi—some once per year while others only every 10 years. Musicians use wooden bachi for practising and ivory bachi only for public performances, although that too varies among musicians. However, one millimetre of wear from the bachi, which touches the strings of the shamisen, significantly reduces sound quality (Tanaka 2008) and musicians need to replace it.

In Tokyo, 29 shops selling shamisen and other Japanese musical instruments were surveyed. The findings:• More than 60% of the shops surveyed knew the

difference between hard and soft ivory. Also, about half the shops stated that hard ivory is the only material useful for bachi.

• Most of the shops did not know the geographical origin of elephant ivory; only 20% mentioned that it came from Africa, but even they had no specific information on areas in Africa.

• Only 10% of the shops had correct information about hard ivory and its origins.It is interesting to note that these shopkeepers were

more familiar with hard ivory than were the ordinary

Nishihara


38% of the ordinary hanko shops (39 of 102 shops) and 45% of the internet hanko shops (15 of 33 shops). This indicates that more than half of the shops are operating illegally without authorization certificates for dealing in ivory.

Relevant ministries have recommended that ivory shops put a CITES seal (the official seal with the CITES logo provided by these Japanese ministries, Fig. 1) on each ivory product sold, with a unique number on the seal to assure clients that the item is legal, each number being registered (JWCS 2000, 2002). However, putting a seal on each ivory product is only a recommendation from the Japanese government, not an obligation. Our investigation indicated that 22% (22 of 102 shops) of ordinary hanko shops did not use seals or understand the system of seals; 73% (24 of 33 shops) of internet hanko shops showed no evidence of using seals; about 80% (23 of 29 shops) of shamisen shops either did not use seals or did not register the numbers that were on the seals. These findings show that shops poorly manage their ivory products, a trend more pronounced in internet hanko shops and shamisen shops than in ordinary hanko shops. An ordinary hanko shop uses the CITES seals but uses the same number on every piece, illegally duplicating the seals and thus making them useless as a proper register.

The department of CITES management at JWRC handles the administration related to ivory registration and certification of ivory products in Japan. Even if Japanese citizens buy (or are given) illegal ivory in Japan, all they need to do is to register such items by using the given format and sending the form to JWRC.

Once certified, the items become legal. During this investigation, one of the officials in JWRC admitted that the system is vague as to how non-registered ivory gains legal certification.

Ivory dealers should know about the CITES regulations in order to manage their ivory business legally. However, their knowledge is poor (JWCS 2000). For instance, we found that less than 20% of ordinary hanko shops knew about CITES and the two legal one-off ivory trades under CITES regulations.

The TIACA staff said that the association is no longer powerful and effective in conducting its responsibilities of creating awareness of the domestic ivory management system and CITES regulations. Younger people are not joining the association and many ivory dealers have left due to the complicated ivory management system. One TIACA member alleged that during the one-off ivory auction in 2008 in South Africa under CITES regulation, hard ivory was found among ivory pieces available in the auction. Japanese dealers selectively bought this ivory, although hard ivory should not have been included in the sale because there are no forest elephants living in southern Africa.

When asked for information, the ministry could not give clear answers regarding 1) the lack of a computer database to manage commodity chains of ivory deals and to easily identify the origin of each ivory product, 2) how to control internet shops, where most dealers do not respect ivory management procedures, 3) why some ivory dealers do not correctly use CITES certificate seals on each ivory product or even have the proper authorization to do so, 4) the current weak

Figure 5. Comparison in size of a 15-kg tusk and a bachi. Although the tusk is long and large, it is almost impossible to make even one bachi from it because such a tusk is usually hollow at the base (as indicated in white) and the maximum diameter of solid ivory is not as wide as the maximum width of bachi (outlined in red).

© E

mik

o N

ishi

hara



legal ivory register system, and 5) the lack of stock quantity and a specific system for managing hard ivory in spite of the strong demand for it in Japan.

There are no strict controls at customs checks in any of the Japanese international airports. For example, Japanese travellers can easily transport pieces of ivory in their luggage. Recently, it was found that some Japanese tourists were bringing in uncarved ivory hanko materials, especially from other Asian countries like China, and having their names carved on them in ordinary hanko shops (JWCS 2000). Four cases were found during this investigation period. These incidents are examples of illegal importation of ivory into Japan and there is no effective system to control this activity. Also, at the Tokyo seaport customs control, it was found that shipping imports controls are lax with no system for determining the geographical origin of confiscated items or to prevent confiscated items from entering local markets.

Discussion

Recommendations to improve the Japanese ivory management systemIn Japan the quantity of musical instrument parts made of ivory is small (Vigne and Martin 2010). More importantly, the focus should be on the size of the tusks necessary to make a complete bachi. In the past, as many as four bachi could be made from large tusks (Martin 1985) weighing at least 15 kg (see Fig. 5 caption). However, data from tusks confiscated by guard patrols in the Republic of Congo (Domingos Dos Santos, pers. comm. 2012) show that tusks weighing more than 15 kg are rare (only one pair weighing 21 kg and two pairs weighing 18 kg among 44 pairs). Tusks from current ivory stocks in Japan weigh on average 12 kg, which is not adequate for even one bachi. If musicians need 100 new bachi during the course of a year, at least 50 forest elephant tusks without cracks and weighing more than 15 kg would be needed, but data from both the field and the Japanese ivory stocks show that it is difficult to meet this demand.

The current Japanese ivory stock management system has no means of quantifying the amount of forest elephant ivory; no differentiation is made between savannah and forest elephants. This is because CITES does not differentiate between the two species of elephants, savannah and forest, although elephants in general are classified as vulnerable under IUCN.

However, recent morphological and genetic studies insist on the distinction that these two so-called sub-species are actually two independent species (Roca et al. 2001, 2007; IUCN/SSC 2002; Rohland et al. 2010; Ishida et al. 2011a, 2011b). This distinction is important because forest elephant populations have drastically declined in number due to poaching (Beyers et al. 2011; Bouche et al. 2011; Maisels et al. in review). There is no hard evidence that ivory originating from forest elephants in Central Africa is smuggled into Japan. However, the question remains whether only old stock of hard ivory can still maintain the demand for bachi that requires large tusks of hard ivory originating from forest elephants in Central Africa. At the least, a system for monitoring stock quantity and a specific management system for hard ivory should be put in place to ascertain the demand for hard ivory in Japan.

The following measures should be improved as mentioned above: 1) a computer database should be created to manage commodity chains of ivory deals and to easily identify the origin of ivory of each ivory product, 2) clear strategies to control internet shops for ivory products, where presently most dealers do not respect ivory management procedures, 3) the regulation needs to be enforced that requires ivory dealers to be authorized and obliges them to use correctly a CITES certificate seal on each ivory product, 4) the current legal ivory register system needs to be improved.

In addition, the relevant ministries should take a strong initiative to create awareness among ivory dealers of the domestic ivory management system and the CITES regulations. At the same time, poor knowledge of elephants and ivory, such as the origin of ivory and the difference between hard and soft ivory, causes a lack of transparency between vendors and clients; more awareness should be created and information disseminated about elephants and ivory.

Japanese customs officials should also establish a system to identify through DNA analysis illegal ivory that was confiscated, in order to determine the quantity of hard ivory illegally destined for Japan from Central Africa.

More investigations needed on hard ivory and its products in Japan

We need additional intensive investigations to learn more about hard ivory and its products in Japan,

Nishihara


especially quantitative data on products. Since bachi is the most numerous item made from hard ivory, it is the most important target to investigate, as the demand for hanko made from ivory has declined. A freelance Japanese journalist, a colleague of the author familiar with Japanese traditional performances such as Kabuki, will conduct interviews and dialogues with staff and performers in the next phase of our study. These participants will range from scholars with expertise in this domain, major managers of performance groups including shamisen players, stage hands and ivory dealers offering bachi to shamisen players. The results will help evaluate the future of traditional and professional demand for hard ivory bachi in Japan. The following questions will be asked:• Who and how many shamisen players need bachi

made of hard ivory?• How many hard ivory tusks of what size are used

annually in making bachi?• What is the existing stock quantity of hard ivory in

Japan available for bachi?• How often do shamisen players replace ivory bachi

with new ones, especially after cracking the tips?• What is the commodity chain from ivory dealers to

bachi makers to musicians?• Are there alternative materials that would suffice

in making bachi?Performers and staff are proud of their traditional

jobs and performances and remain in an isolated setting with little outside communication. They are likely unaware of the status of wildlife in Central Africa or the role of the ivory trade. Given that ivory-made bachi are essential to their performances, discussions will revolve around the theme of how to find a balance between traditional Japanese cultural values and biodiversity conservation. The initial meeting will gather information to determine the existing knowledge of these performers and their use of ivory, and introduce them to forest elephants and their conservation status. The second meeting will be a follow-up with the same group to gauge how views and practices of staff have changed and to focus more on conservation issues. These meetings will use educational materials (see below).

More awareness needed in Japan on elephant conservation

Traditionally, Japan has a culture that respects nature and wildlife. The Japanese enjoy seasonal displays of

spring blossoms, summer green forests, autumn leaves and winter landscapes. They eat little wild bush meat. These customs and appreciation of nature and wildlife are expressed in Japanese traditional poems (waka or haiku) and in Japanese folklore where many wild animals appear as friendly neighbours. The Japanese people also enjoy imitating nature—gardening, flower arrangement, bonsai, etc.

Because of this traditional Japanese concept of nature and wildlife, it can be said that historically the Japanese have a good sense of nature and wildlife conservation in Japan. However, these trends have almost no relationship with the fact that Japanese people do not fully understand the importance of elephant conservation. Indeed, the Japanese have been using ivory without a conservation concept. This is mainly because the qualities of ivory—its colour, durability, absorption properties and hardness—allow it to be carved into detail. It is also partly related to particular Japanese Shinto spirits, which encourage using natural materials such as ivory as they are the best nature spirits (Motegi 1988; Takeda 2010). The demand for ivory for hanko production is currently declining, because of the psychological pressure of the ivory ban and also simply because the younger generation is losing interest in luxury ivory products (Vigne and Martin 2010), and not because the Japanese comprehend global conservation.

The relationship between traditional culture in Japan and ivory use is also vague. The use of hanko to identify any documents needing certification was widespread in the Edo era, more than 300 years ago. Originally, major materials for hanko were stones of clear crystals (http://www.rokugo.com). Hanko culture and carving techniques are traditional with a long history, but not for hanko made of ivory, a practice that came into fashion just after World War II. Shamisen and bachi are also part of traditional culture but not bachi made of ivory. Shamisen dates back more than 500 years, and this instrument has been used in well-known Japanese performances with 400 years of history. But bachi made of ivory became popular only after the Meiji era, about 150 years ago. Thus, we suggest that the Japanese people should be taught to understand that using ivory is not part of traditional culture but is used for technical reasons.

In Japan, one urgent task is to establish an information-sharing mechanism on wildlife conservation, particularly on issues outside Japan. The Japanese need to understand in the context of



global biodiversity conservation the adverse effects of using wildlife products. In Japan, there is neither much opportunity nor material from which to learn. Especially, little information is available about the African tropical forest and its conservation status, including about forest elephants, their ecology and the threats to their existence.

It is essential to create educational materials for use in Japan aimed at many and varied target groups: schools, universities, zoos, non-government organizations (NGOs), businesses involved in corporate social responsibility, ecotourism sectors, ivory dealers, traditional artists such as Kabuki actors and shamisen players, ministries, development people, and also for the internet. The author initiated a project in April 2012 with a Japanese NGO and a professional Japanese educator in biodiversity conservation who visited the Central African region to make conservation education materials under the author’s supervision. These materials will be distributed widely to the Japanese public. Also, field training sessions for young Japanese in the Central African forest area, an area of the world unfamiliar to most Japanese, should be organized. These field trainees who are interested in careers in conservation will become not only future field conservationists but also strong messengers spreading the global conservation ethic to the Japanese. The author is in the ideal position to initiate this action, as the only Japanese who has been working in Central African conservation under the Wildlife Conservation Society (WCS) for more than two decades.

China’s ivory demand compared with that of Japan

After the CITES ban of 1989, Japan was allowed a one-off ivory trade from southern African nations in 1999 under CITES regulations. A second similar shipment was allowed in 2008 for both Japan and China. China has more than 1,800 years of history using ivory. The demand for ivory in China is growing because it has a population of more than a billion people, and growing middle and upper classes that buy ivory. At the same time it should be noted that currently more and more Chinese people in Africa are buying raw tusks and ivory products and are linked to illegal smuggling operations. This situation has resulted in serious poaching pressure on elephants and the illicit trade of ivory everywhere, particularly between Africa and

China (Gabriel et al. 2012).Such Chinese demands need to be expressed in

relation to Japan’s demand in order to compare and contrast the trends of the two countries. In the past the Chinese made personal seals out of ivory, but the current trend is mostly to use ivory in crafts and accessories (Gabriel et al. 2012), using any quality of ivory including cracked portions of tusks (Gendai Inshou 2009). This is related to the current Chinese boom of using mammoth tusks, most of which are cracked (Martin & Vigne 2011) and are considered of lower quality by Japanese ivory dealers. Also, the Chinese do not differentiate between hard and soft ivory (Yan Xie, pers. comm. 2010). Japan bought expensive hard ivory at the CITES ivory auction in 2008 without competition from Chinese buyers, who bought the cheaper, softer ivory from savannah elephants, including cracked tusks. This indicates that Japan has a specific demand for hard and non-cracked ivory.

Conclusions• Japan is the only country in the world where a

strong demand for hard ivory, originating from forest elephants in Central Africa region, still exists.

• Though there are no quantitative data, currently most of the demand for hard ivory in Japan is for bachi, used in playing the Japanese traditional musical instrument, the shamisen, with a decline in ivory-made hanko.

• One bachi requires one non-cracked complete tusk of more than 15 kg, and Japanese shamisen players need to replace bachi once the edge is cracked.

• We need to determine if old stocks of hard ivory are available to match the demand for bachi, which require a large tusk for each piece.

• Most Japanese ivory dealers have insufficient knowledge of the domestic management system for ivory, CITES regulations, and general information about elephants and ivory, which contributes to illegal ivory imports into Japan.

• It is recommended that the current Japanese man-agement system for ivory be improved, especially to enforce control of hard ivory, for which there is great demand.

• More investigations on the use of hard ivory, espe-cially for bachi, are planned to get more quantita-tive data.

• As traditional Japanese culture appreciates nature

Nishihara


and wildlife, it is important to produce educational materials on global biodiversity conservation that will make the Japanese public aware of the conser-vation issues concerning elephants.

• Japan and China confront different challenges re-garding the potential for illegal ivory to come into their countries and their responsibility in a large part for future elephant conservation.

AcknowledgementsThis report is based on information gathered under the Toyota research grant project November 2009–October 2011, ‘Re-evaluation of value of Japanese traditional culture using ivory and study of forest elephant poaching in Central Africa’ (D09-R-0129), led by Tomoaki Nishihara, under the WCS Congo program. Many thanks to WCS, particularly the WCS Congo program director, Dr Paul T. Telfer, who has strongly supported this project from the beginning. As my principal job is in the field in northern Congo, I am deeply indebted to five Japanese investigators—Fumiaki Nagaishi, Hideaki Fujii, Tamiko Tamura, Yoshio Katsui and Yukiyo Enomoto, and the coordinator of those investigators, Keiko Ikemoto—who helped in this project. I wish to thank Andrea Turkalo, who helped revise this paper and encouraged its publication, and to Emiko Nishihara, who produced Figure 5.

References

Anon. 1690. Jin Rin Kin Mou Zui. Asakura, H. (ed.). Touyou Bunko 519, Heibon-sha (in Japanese).

Association of Japanese Actors. 2000. Performance art: its technicians at Kabuki. Yagi Publisher (in Japanese).

Beyers, R., Sinclair, T., Hart, J., Grossman, F., Dino, S. and Klinlenberg, B. 2011. Resource wars and conflict ivory. The impact of civil conflict on elephants in the Okapi Faunal Reserve: 1995–2006. PLoS One 6(11):e27129. doi:10.1371/journal.pone.0027129.

Bouche, P., Douglas-Hamilton, I., Wittemyer, G., Nianogo, A.J., Doucet, J.L., Lejeune, P. and Vermeulen, C. 2011. Will elephants soon disappear from West African savannahs? PLoS One 6:1–11.

Gabriel, G., Hua, N. and Wang, J. 2012. Making a killing: A 2011 survey of ivory markets in China. International Fund for Animal Welfare.

Gendai Inshou. 2009. June 2009. 12 pp. (in Japanese).

Ishida, Y., Demeke, Y., Van Coeverden de Groot, P.J., Georgiadis, N.J., Leggett, K.E.A., Fox, V.E. and Roca, A.L. 2011a. Distinguishing forest and savannah African elephants using short nuclear DNA sequences. Journal of Heredity 102(5):610–616.

Ishida, Y., Oleksyk, T.K., Georgiadis, N.J., David, V.A., Zhao, K., Stephens, R.M., Kolokotronis, S.O. and Roca, A.L. 2011b. Reconciling apparent conflicts between mitochondrial and nuclear phylogenies in African elephants. PLoS ONE 6:1–16.

IUCN/SSC African Elephant Specialist Group. 2002. Statement of the taxonomy of extant Loxodonta, February 2002.

[JWCS] Japan Wildlife Conservation Society. 2000. Effect of resumption of international trade on Japanese ivory market. Tokyo: JWCS.

[JWCS] Japan Wildlife Conservation Society. 2002. Black and grey illegal ivory and Japanese markets. Tokyo: JWCS.

[JWRC] Japan Wildlife Research Center. 2007. Report on evaluation of national ivory-trade management system and investigation for its improvement (in Japanese). Tokyo: JWRC.

Maisels F. et al. in review. Devastating decline of forest elephants in Central Africa. PLoS One.

Martin, E. 1985. The Japanese ivory industry. Tokyo: World Wildlife Fund.

Martin, E. and Stiles, D. 2003. The ivory markets of East Asia. London and Nairobi: Save the Elephants.

Martin, E. and Vigne, L. 2011. The ivory dynasty: a report on the soaring demand for elephant and mammoth ivory in southern China. London: Elephant Family, Aspinall Foundation, Columbus Zoo and Aquarium.

Motegi, K. 1988. Nihon no Gakki – sono sozai to hibiki (Japanese music instrument – its materials and sound). Ongaku-no-Tomo sha (in Japanese).

Nishihara, T. 2003. Elephant poaching and ivory trafficking in African tropical forest with special reference to the Republic of Congo. Pachyderm 34:66–74.

Roca, A.L., Georgiadis, N., Pecon-Slanttery, J. and O’Brien, S.J. 2001. Genetic evidence for two species of elephant in Africa. Science 293:1473–1477.

Roca, A.L., Georgiadis, N., Pecon-Slanttery, J. and O’Brien, S.J. 2007. Cyto-nuclear genomic dissociation and the African elephant species question. Science Direct Quaternary International 169–170:4–16.

Rohland, N., Reich, D., Mallick, S., Meyer, M., Green, R.E., Georgiadis, N.J., Roca, A.L. and Hofreiter, M. 2010. Genomic DNA sequences from mastodon and



woolly mammoth reveal deep speciation of forest and savannah elephants. PLoS Biology 8:1–10.

Takeda, T. 2010. Why the Japanese people are the most attractive in the world. PHP Shinsho 705 (in Japanese).

Tanabe, H. 1963. Shamisen Ongakushi (history of shamisen). Soushisha (in Japanese).

Tanaka, K. 2008. Zukai Nihon Ongakushi (Japanese history of music with pictures). Tokyo-do shuppan (in Japanese).

Tanaka, Y., Nogawa, M. and Haikawa, M. 2009. Marugoto Shamisen no hon (Shamisen, a Japanese

traditional music instrument). Seikyuu-sha (in Japanese).

TRAFFIC. 2011. ‘Annus horribilis’ for African elephants. TRAFFIC press release, 29 December 2011.

Tsugawa, N. 2004. Shamisen wo hajimeyou (Let’s start playing shamisen, a Japanese traditional music instrument). Seibi-do (in Japanese).

Vigne, L. and Martin, E. 2010. Consumer demand for ivory in Japan declines. Pachyderm 47:45–54.


Patton et al.

Dispersal and social behaviour of the three adult female white rhinos at Ziwa Rhino Sanctuary in the immediate period before, during and after calving

Felix J. Patton,* Petra E Campbell, Angie Genade, Robert Ayiko and Godfrey Lutalo

Ziwa Rhino Sanctuary, Nagasongola, Uganda*corresponding author email: [email protected]

AbstractThe paper presents details of the dispersal and social behaviour of the three adult female white rhinos at Ziwa Rhino Sanctuary in Uganda in the period one month before, the month during and one month after calving where published information from other reserves is limited. All six births occurred during night-time hours in areas of dense habitat. Births took place within 32 hours of the older calves being chased away by their mothers. Each female moved within the rhinos’ preferred habitat of open woodland with short grass and spent most of her time alone with her new calf and little time associating with other rhinos in the population. Apart from the chasing away of the older calves, there was no indicator of an impending birth.

RésuméLe document donne les détails de la dispersion et du comportement social des trois rhinocéros blancs femelles adultes au sanctuaire de Ziwa au cours du mois avant, pendant le mois et le mois après le vêlage là où les informations publiées à partir d’autres réserves sont limités. Les six naissances ont eu lieu durant la nuit dans des zone d’habitat dense. Les naissances ont eu lieu dans les 32 heures après que les jeunes rhinocéros plus âgés aient été chassés par leurs mères. Chaque femelle s’est déplacée vers l’habitat de rhinocéros préféré de fôrets claires aux herbes courtes où elle a passé la plupart de son temps seule avec son nouveau bébé en s’associant peu avec d’autres rhinocéros dans la population. En dehors du fait d’avoir chassé des jeunes rhinocéros plus âgés, il n’y avait aucun autre indicateur d’une naissance imminente.

IntroductionA search of the published literature on the behaviour of white rhino females and their associates at and around the birth of a new calf showed that there was a paucity of detailed information on the subject. Owen-Smith (1975) is the research most often referred to in papers relevant to the social behaviour of white rhinos. He reported that calves were driven away by their mothers prior to the birth of a subsequent calf, but gave no details as to how soon after the former calf was driven away the new birth took place or at what time of day or in what type of habitat births took place.

No reports could be found to indicate whether the behaviour of the female in the weeks leading up to the birth of a new calf altered in any way that could be

used as an indicator of an impending birth. Such an indicator may be useful in the management of the cow and calf where additional security may be required to ensure their survival.

Limited information exists on the changes to the social behaviour of a white rhino population after the birth of calves. Owen-Smith (1975) reported that a newly released calf might be found with a mother–offspring pair but that such associations were mostly temporary and the pair was not related to it. Owen-Smith also stated that most subadults not accompanying cows were associated with a similarly aged companion of the same or opposite sex with a group structure of up to five individuals. Shrader and Owen-Smith (2002) concluded that new subadults prefer to seek an association rather than remain alone.


Behaviour of three white rhinos in the period before, during and after calving

This paper presents a detailed analysis of the dispersal and social behaviour of the three adult female white rhinos at Ziwa Rhino Sanctuary in the period one month before, the month during and one month after the three females calved. Published information from other reserves is limited. The study identifies the time of day, location and habitat when the rhinos calved, the length of time between the older calf being chased away and the new birth, whether the females changed their location or habitat immediately before and immediately after calving, and whether there was a noticeable effect on the females’ sociability (associations) with the other rhinos in the population.

A principle objective of the study was to determine if the females made any noticeable changes within the few days leading up to the birth of their new calves that reserve managers could use as indicators of an impending birth.

MethodZiwa Rhino Sanctuary cov-ers 64.2 km2 in Nagasongo-la District, central Uganda. Before becoming a wildlife sanctuary in 2004 the area was part of a cattle farming operation. About 30% of the reserve is swamp, but only the southern swamp area is always water bound. Other swamp areas dry out in times of low rainfall and flood during high rainfall.

For security and monitoring purposes, the reserve is organized into five sectors: Karakwende (K), Mikerenge (M), Wangoriro (W), Rwanyanya (R) and Lugogo (L). Each sector is further organized into either three or four numbered blocks. Figure 1 shows the location of each of the sectors and blocks.

At the start of 2011, the white rhino population of Ziwa Rhino Sanctuary consisted of three adult

males—Taleo, Moja and Hassani—and three adult females with calves—Nandi and calf Obama, Bella and calf Augustu, Kori and calf Justus. The calves were the first-borns of the three females. In June 2011, Nandi produced a second calf, Malaika, while Bella had Donna and Kori had Laloyo in January 2012.

The rhinos at Ziwa are held under heavy 24-hour-a-day security by armed guards and monitoring rangers. Since June 2010 to better understand the behaviour and movements of the rhinos the monitors have kept an hourly record of the location, key activities of each rhino and their associations with conspecifics (for more details see Patton et al. 2011).

Rangers monitoring each of the Ziwa rhinos—six adults and six calves—follow the animals on foot from a distance but keeping them in sight as much as possible. During the hours of darkness, rangers have torches, which they use to observe the rhinos whenever they hear any sound of activity, and at least at 15-minute intervals to check on the rhinos. During full moon periods the rhinos are easily observed without the need for torches.

dense bushland

W1

W2

W3

R1 M2

R3

L3

K1

W = Wangoriro

L = Lugogo

R = Rwanyanya

M = Mikerenge

K = Karakwende/Kamira/ Kasozi

R2

L2

L1

open woodland with short grass and a few thickets

dense bushland with tall and short grass

belt of tall grass dominated by loudetia, digitaria, with shrubsrhus, acacia and phoenix (palm)

short grassland with scattered trees

belt of short grass, cyperus and digitaria

airstrip, tall grass of hyperrhenia dominated by Acacia hockii

belt of swamp with papyrusswamp consists mainly of perotis, brachiaria, digitaria and hyparrhenia grasses

Figure 1. Basic vegetation map of Ziwa Rhino Sanctuary.


Patton et al.

The rangers completed a daily sighting form for each hour of the 24 hours of each day of each of the months from June 2010 to May 2012. For each rhino the rangers recorded the location—sector and block; type of habitat—wet swamp, dry swamp or other; one of three main activities—feeding, resting, moving; plus any of four secondary activities—drinking, wallowing, mating or fighting—that might also have occurred during each hour. All other rhinos found in close proximity (within 10 m) were recorded as an association.

Data were entered using Microsoft Excel 2007 on a Toshiba Satellite Pro laptop. Macros were written to automate the consolidation and analysis of the data. For location and habitat type, the analysis was based on the number of rhinos found in each location and habitat type for each hour of the day; for activities the analysis was based on the number of hours spent on the activity per day. The daily data were not always complete as there were occasions when the particular rhino was out of sight (especially in thick bush habitat) or had run off and had to be found again, especially after a disturbance at night.

Data analysis was carried out for female 1 (Nandi) for the period May–July 2011, and females 2 (Bella) and 3 (Kori) for the period December 2011–February 2012. The number of hours between the time the female chased the older calf away and gave birth to the new calf was determined along with the location and time of birth of each new calf. The number of hours and percentage of time that each female spent in each sector in the period before and after calving and the difference between the two were considered. The associations between the three adult females with the other rhinos in the population before and after the birth of their new calves were analysed by the number of hours and the number of events where an association took place in the relevant time periods.

ResultsTable 1 shows details of the birth date, location and time of day of the six calves born at Ziwa. All births

occurred during night-time but in different sectors and blocks. Nandi pushed away her first calf, Obama, around 32 hours before her new calf, Malaika, was born. Bella pushed away her first calf, Augustu, around 8 hours before she gave birth to her new calf, Donna. Bella started getting aggressive with Augustu 3 hours before he finally parted. Kori’s first calf, Justus, left Kori without aggression (together with Obama and Augustu) around 3 hours before her new calf, Laloyo, was born.

The three females moved independently of each other. Table 2 shows the total number of hours analysed for each female, the number of major changes of location made during these hours and the average number of hours per change in location for the research period. A major location change was determined when the rhino was recorded in a different sector or block after an hour.

The results show that the amount of data collected for Kori (total hours) is marginally outside the range of the standard deviation for the group with a consequent similarly marginal effect on the analysis of location changes and average hours per location. However, given the small number of individuals and variability in the efficiency of some rangers in maintaining contact with the monitored rhinos, the data collected can

Table 1. Data on birth of the six white rhino calves born at Ziwa

Calf name Birth date Birth location

Time of birth*

Obama 24 June 2009 R1/M2 in the early hours a.m.Augustu 07 Oct 2009 R3 in the early hours a.m.Justus 02 Jan 2010 L2 believed around 2 a.m.Malaika 04 June 2011 L2 between 11 p.m. and 12

a.m.Donna 10 Jan 2012 W2 around 11 p.m.Laloyo 15 Jan 2012 R2 between 12 a.m. and 5 a.m.

*All births occurred at night.

Table 2. Summary of location changes made by the Ziwa females in the research periodName Total hours

(no.)Location changes (no.)

Average hours per location change (no.)

Bella 2,142 136 15.7Kori 1,832 107 17.1Nandi 2,188 135 16.2Mean 2,054 126 16.3SD 194 17 0.71



be considered sufficiently consistent to individually compare the results of the three females.

All three females changed their preferred location after calving (Table 3). Nandi spent less time in sector L (–22%) and more time in sector W (+32%); Bella spent less time in sector W (–27%) and more time in sector L (+34%); Kori spent less time in sector W (–49%) and more time in sector L (+65%). The main vegetation type in both the sectors L and W is similar, open woodland with short grassland and dense thickets, but sector W is on higher ground than sector L.

The three females did not change their locations to calve, but they did move to a neighbouring sector/block: after 69 hours for Nandi, after 215 hours for Bella and after 288 hours for Kori.

Table 4 shows the effect of the birth of the new calf on the sociability of the three females with other individuals in the population and the immediate effect of the mothers releasing the three older calves on the social organization of the population.

The data show that the total length of associations

and the total number of events when an association took place were greatly reduced after calving for all three females. Obama remained in association with his mother Nandi for only 191 hours while Augustu and Justus had no association with their mothers at any time in the month following the new calvings.

DiscussionThe white rhino mother chasing away her older calf was the only indication of an impending birth in the wild. Our study showed that the female chased away the older calf between 3 and 32 hours before the birth of the new calf. Even the maximum of 32 hours gives little prior warning that the birth of a new calf is imminent.

Location

The females made no major location change around the time of calving, during which the rhino was recorded in a different sector or block in the hour

Table 4. Associations of the female rhinos with other rhinos at Ziwa before and after their calving, shown as total hours/no. of events

Nandi Bella Kori

 Association Before After Diff. Before After Diff. Before After Diff.

All associationsMales Taleo Moja HassaniFirst calves Obama Augustu JustusFemales Nandi Bella Kori

287/32

81/936/691/9

ALL0/00/0

-27/352/5

178/16*

73/514/291/9

191/10/00/0

-0/00/0

–109/–16 

–8/–4–22/–4 

0/0 

191/10/0 0/0 

  -

–27/–3–52/–5

250/18

34/285/555/6

52/2ALL0/0

0/0 -

24/3

128/12*

38/412/253/4

0/00/00/0

9/1-

16/1

–122/–6

4/2 –73/–3 

–2/–2  

–52/–2 n/a 0/0 

 9/1 

 -–8/–2 

1194/18

0/040/418/1

953/763/1ALL

4/153/3

 –

41/7*

0/017/3

8/2

0/00/00/0

0/016/1

 –

–1153/–11 

0/0 –23/–1

–10/1

 –953/–7–63/–1

n/a 

 –4/–1 –37/–2

 –* not including the association with first calfALL represents the first calf was always associated with its mother at all times up to the birth of the new calf.

Table 3. Location of the female rhinos at Ziwa in the period around their calving date

  Nandi Bella Kori

Block Before After Difference Before After Difference Before After Difference

KLMRW

45 (0)*590 (77)

6 (1)104 (14)

16 (2)

0 (0)674 (55)

0 (0)129 (10)417 (34)

–45 (0)84 (–22)

–6 (–1)25 (–4)

401 (32)

7 (0)31 (3)25 (3)

155 (17)691 (76)

0 (0)441 (37)

0 (0)165 (14)593 (49)

–7 (0)410 (34)–25 (–3)10 (–3)

–98 (–27)

0 (0)139 (13)

0 (0)274 (25)671 (62)

0 (0)867 (78)

0 (0)93 (8)

144 (13)

0 (0)728 (65)

0 (0)–181 (–17)–527 (–49)

Total 761 1220 909 1199 1078 1104

* Figures in parentheses are percentage of the equivalent time.


Patton et al.

before. But all births took place in dense habitat and out of direct sight of the monitoring team. However, the females with their new calves moved into new areas relatively soon (3 days, 9 days and 12 days) after the births. Both before and after the new births, the females’ locations were in open woodland with short grass and a few thickets. Owen-Smith (1975) and Schrader (2003) reported that white rhinos primarily fed on short to intermediate-height swards of green grass. This was similar to the finding at Ziwa but with the thickets allowing the females to calve in a more secluded and secure habitat.

The females changing their location shortly after the birth of their new calves can be explained by the effect of the amount of rainfall in the relevant months on the level of waterlogging of the soil in the former and new locations. Sector L is in close proximity to the permanent swamp and has a high water table such that the ground gets waterlogged after rain. Sector W is on higher ground in the north-west of the reserve, which tends to dry out during periods of limited rain.

Before Nandi calved in June 2011, there had been heavy rainfall in May 2011, as there was in July. When accompanied by her new calf, Nandi showed a preference for sector W where the higher ground was drier than in sector L where she had stayed with her first calf.

The opposite was recorded for Bella and Kori when accompanied by their new calves. December 2011, January and February 2012 were all low rainfall months. The mothers and calves moved from the higher, drier ground of sector W to the lower ground of sector L where water was available despite the dry months.

Whether the presence of the new calf precipitated an earlier move in location is open to conjecture. On one hand, a newly born calf may be considered to be more susceptible to disease if constantly standing and lying in wet conditions, while on the other hand it is likely that a white rhino mother with a new calf at foot will need a higher than normal intake of water to maintain her milk supply. The mother may react to the presence of too much or too little water and change location accordingly.

With the knowledge that major changes in the chosen locations some days after the births may occur, security and monitoring may be enhanced at this time to ensure the maintenance of efficient and effective systems.

Associations

All three females made fewer associations in the period immediately after the birth of their new calves than in the period immediately before.

Nandi associated strongly with the two males Hassani and Taleo both before and after she calved. Bella and Kori associated mostly with the male Moja. The association between Nandi and Hassani was considered to be non-sexual with no mating behaviour observed. The two were raised at the same location in the USA before they were translocated to Ziwa. Taleo was observed to have mated with Nandi, producing her first offspring, Obama, and was probably the parent of the new calf (subject to proofing by DNA evidence). The similarity in the amount of time Taleo spent in association with Nandi both before and after the female calved supports this conclusion.

The association between Kori and Moja was observed to be sexual with mating behaviour previously having been recorded. The absence of the other breeding male, Taleo, before and after the female calved and the presence of Moja indicate that Moja was most likely the parent of the new calf (subject to proofing by DNA evidence).

Obama was with his mother, Nandi, up to the time of the new birth. After the calving, Obama remained in close association with his mother for nearly 8 days (191 hours). He then moved away to form an association for 953 hours with the female Kori and her calf Justus up until the birth of Kori’s new calf. Owen-Smith (1975) reported that a newly released calf might be found with a mother–offspring pair but that such associations were mostly temporary and the pair were not related to the calf. This was the situation recorded for Obama but it should be noted that at this time there was no alternative association to a mother–offspring pair as there were no other subadults in the population for Obama to associate with.

Augustu was in association with his mother, Bella, until the birth of the new calf but remained for only one day afterwards while Justus ceased association with his mother, Kori, the day before the birth of the new calf.

The three first calves joined up in association from 15 January, just a few hours after the final birth, that of Kori, and remained together for the rest of the reporting period. This form of association was also reported by Owen-Smith (1975), who stated that most subadults not accompanying cows were associated with a similarly aged companion of the



same or opposite sex with a group structure of up to five individuals. Owen-Smith regarded associations of more than two subadults to be unstable.

Shrader and Owen-Smith (2002) concluded that new subadults prefer to associate rather than remain alone. This was considered a security measure against the adult males. Owen-Smith (1974) reported an instance where a solitary subadult male was challenged tensely for 32 minutes by a territorial male. However, another territorial male had confronted this same subadult only briefly 5 months earlier, when the latter had another subadult male companion. Taleo, the dominant male in the Ziwa population, was found to be particularly aggressive towards any unaccompanied subadult.

ConclusionThis paper presents details of the dispersal and social behaviour of three adult female white rhinos in the immediate period surrounding calving, a subject about which little has been published. A principle objective of the study was to determine if there were any noticeable behaviour changes of the females within the few days leading up to the birth of their new calves which could be used by reserve managers as indicators of an impending birth, for example, a noticeable shift in location or the older calf being found alone having been chased away by the mother.

The Ziwa females showed no change in location before the birth of the new calves. Each female remained with her current calf almost until the birth of the new calf. As such, there was no behavioural change that acted as an indicator of the impending calving.

Acknowledgement

The authors acknowledge the work of the Ziwa rhino monitoring rangers in collecting the daily rhino behaviour data, without which the analysis would not have been possible.

ReferencesOwen-Smith, N. 1974. The social system of the white

rhinoceros. In: Geist, V. and Walther F.R. (eds.), The behaviour of ungulates and its relation to management, IUCN, Morges. pp. 341–351.

Owen-Smith, N. 1975. The social ethology of the white rhinoceros Ceratotherium simum (Burchell 1817). Zeitschrift für Tierpsychologie 38:337–384.

Patton, F.J., Campbell, P.E., Genade, A., Ayiko, R. and Lutalo, G. 2011. The behaviour of white rhinos at Ziwa Rhino Sanctuary, Uganda, with particular reference to night-time activity. Pachyderm 50:77–83.

Schrader, A.M. 2003. Use of food and space by white rhinos. PhD thesis. University of Witwatersrand, Johannesburg, South Africa.

Shrader, A.M. and Owen-Smith, N. 2002. The role of companionship in the dispersal of white rhinoceroses (Ceratotherium simum). Behavioral Ecology and Sociobiology 52:255–261.


Jachmann

Pilot study to validate PIKE-based inferences at site level

Hugo Jachmann

Bergstaat 77, 6174RP Sweikhuizen, The Netherlandsemail: [email protected]

AbstractThe primary objective of the Monitoring of Illegal Killing of Elephants (MIKE) programme is to monitor worldwide trends in elephant poaching. MIKE has been employing in its analyses the proportion of illegally killed elephants (PIKE) as a relative indicator of poaching levels. PIKE is subject to a number of potential biases that need to be understood to assess the validity of inferences made from analyses based on it. In four well-managed sites (Queen Elizabeth and Murchison Falls in Uganda, and Mole and Kakum in Ghana), a pilot study was carried out to examine the on-site reliability of PIKE. The detection probability of elephant carcasses was examined in relation to visibility (habitat types) on patrol, and PIKE results were compared with the results of a catch per unit effort (C/E) analysis. Due to sharply increased patrol coverage in three out of four sites, by 2011, the detection probability of elephant carcasses approached 1. PIKE, based on verified field data corrected for detection probability, compared well with data from the carcass sheets in the MIKE database. C/E results provided support for the on-site use of PIKE. Because the relationship between the C/E index and law-enforcement effort varied by site, which makes pooling of data complicated, for MIKE’s purposes C/E analysis was not considered a practical tool with which to monitor worldwide elephant poaching. PIKE data on the other hand can be easily pooled, and the pooling may erode away some of its imperfections. PIKE was found to be superior to C/E analysis for a system like MIKE.

Additional key words: MIKE, elephant poaching, monitoring

RésuméL’objectif principal du programme du Suivi de l’abattage illégal des éléphants (MIKE) est de faire le suivi des tendances mondiales du braconnage des éléphants. Dans ses analyses MIKE utilise la proportion des éléphants abattus illégalement (PIKE) comme un indicateur relatif du niveau de braconnage. PIKE est sujet à un certain nombre de partis pris potentiels. On doit comprendre ces partis pris afin d’évaluer la validité des conclusions faites à partir des analyses qui se basent sur PIKE. Sur quatre sites bien gérés (Queen Elizabeth et Murchison Falls en Ouganda, Mole et Kakum au Ghana), une étude pilote a été réalisée pour examiner la fiabilité de PIKE sur le site. La probabilité de détection des carcasses d’éléphants a été examinée par rapport à la visibilité (types d’habitats) en patrouille, et on a comparé les résultats de PIKE à ceux de l’analyse de la capture par unité d’effort (C/E). En raison de la forte intensification de la couverture de patrouille dans trois des quatre sites, jusqu’en 2011, la probabilité de détection des carcasses d’éléphants avoisinait 1. PIKE basée sur des données vérifiées, corrigée pour la probabilité de détection, s’est comparée bien aux fiches de carcasse dans la base de données de MIKE. Les résultats de l’analyse C/E ont appuyé l’utilisation de PIKE sur le site. Puisque la relation entre l’indice de l’analyse de C/E et l’effort de l’application de la loi variait selon le site, ce qui complique la mise en commun des données, pour les fins de MIKE, on ne considère pas l’analyse C/E d’être un outil pratique

MANAGEMENT



Introduction

The Monitoring of Illegal Killing of Elephants (MIKE) programme, approved in 1997 by the Convention on International Trade in Endangered Species (CITES) and ratified at the 11th Convention of Parties in 2000, was set up to monitor trends in elephant poaching so that decisions on elephant issues taken by CITES are based on sound information. Poaching trends are monitored in a sample of about 80 sites spread across the range of African and Asian elephants. The second objective of MIKE is to build capacity in range States to manage elephant populations. The MIKE programme was designed to analyse data on elephant mortality and law-enforcement effort, obtained primarily from law-enforcement patrols local patrol staff routinely conduct at designated MIKE sites. While a reasonable amount of data on elephant mortality has been compiled since MIKE’s inception in 1997, it has not been possible to obtain law-enforcement effort data from the vast majority of MIKE sites, primarily due to a number of operational difficulties.

In 2008 the MIKE programme initiated the deployment of MIST (Management Information System), a comprehensive field data management information system. It was anticipated that by better serving the protected area management needs of range states (and not just those of MIKE), MIST would help to improve the quantity and quality of data flowing to MIKE. Although the goal of obtaining comprehensive law-enforcement data maintains a high priority on the agenda for MIKE, and while MIST deployment has been progressing well, most sites are not yet ready to deliver effort data.

In view of the lack of effort data, and to meet its reporting obligations to the CITES parties, the MIKE programme has been employing in its analyses the proportion of illegally killed elephants (PIKE) as a relative indicator of poaching levels. PIKE is the number of illegally killed elephants found divided by the total number of elephant carcasses encountered by patrols or other means, aggregated by year for each site. PIKE is a relative indicator of elephant poaching

levels that does not require adjustment by level of law-enforcement effort but is subject to a number of potential biases. These biases need to be understood in order to assess the validity of inferences made from analyses based on PIKE.

This paper presents the results of a pilot study carried out from March to April 2012 to examine the reliability of the within-site sample used to compute PIKE. Mainly due to time constraints, this pilot study was not expected to result in conclusive data on PIKE’s reliability, but to provide the foundation of a comprehensive study planned for phase 3 of the MIKE programme.

We began by looking at the detection probability of elephant carcasses in relation to visibility or habitat type, to compare PIKE calculated from the raw carcass data with that corrected for differential detection probabilities. We then examined the detection probability in relation to the cause of death, e.g. natural deaths, illegally killed, and those in the unknown category. Because law-enforcement effort data were available for the four sites that were selected for this study, we compared PIKE with the results of a catch per unit effort analysis (C/E), first to look at the pitfalls and merits of both methods, and second to examine whether C/E results provided support for PIKE-based inferences at the site level. We concluded by comparing PIKE with absolute elephant mortality.

Study areas

For this pilot study we focused on relatively small populations in relatively small but well-managed sites (≤ 4,500 km²), representative of the main habitat types found within the elephant range (savannah and forest), with MIST as the main tool for monitoring law enforcement. Considering these broad criteria, we selected four sites:1. Kakum Conservation Area (CA) in Ghana

comprises Kakum National Park (NP) and the adjoining Assin Attandanso Resource Reserve, both covered by moist evergreen forest of the Upper Guinea forest belt, totalling 366 km². The last count

pour surveiller le braconnage des éléphants dans le monde entier. D’autre part, on peut facilement mettre en commun les données de PIKE, et la mise en commun peut éroder certaines de ses imperfections. On a donc trouvé que PIKE était meilleure que l’analyse C/E pour un système comme MIKE.

Mots clés supplémentaires: MIKE, braconnage des éléphants, monitoring


Jachmann

in 2004 returned 164 elephants (Danquah 2004).2. Mole NP in Ghana is covered by Guinea savannah

woodland with gallery forests along the main rivers; it totals 4,504 km². The last count in 2006 returned 401 elephants (Bouché 2007).

3. Murchison Falls NP in Uganda comprises open savannah and woodland, with gallery forest along the main rivers, totalling 3,893 km². The last count in 2010 returned 904 elephants (Rwetsiba & Wanyama 2010).

4. Queen Elizabeth NP in Uganda has a mosaic of open savannah, wetlands and semi-deciduous forest, covering 2,294 km². The last count in 2010 returned 2,502 elephants (Plumptre et al. 2010).

Methods and materials

Detection probabilityIn each of the four sites, we estimated the maximum mean strip width with regard to detecting elephant carcasses. Using a vegetation map combined with maps of the road system and elephant distribution, we applied a simple general stratification process based on three factors: elephant distribution, ratio of dominant vegetation types, and presence of roads. Layout of transects followed elephant distribution and was in proportion to the distribution of dominant vegetation types, but due to time constraints accessibility (road network) remained an important factor. Each of the roads selected was sampled both left and right at exactly 1-km intervals for the three savannah sites, and at 50-m intervals at Kakum, by sending a patrol staff at a perpendicular walk away from the road until roughly 1.3 m of the person became invisible to the observers (only head and shoulders visible). However, this approach may result in a biased estimate of strip width: first, because a moving object is easier to detect than a stationary one, and second, because an object moving away from focused observers may be visible longer than stationary objects at shorter distances. These sources of bias are interrelated and may lead to an overestimate of strip width and an underestimate of carcass densities. There was simply not enough time to carry out a more complicated, bias-free design using stationary objects. Moreover, carcasses in an advanced state of decay, such as merely bones and some skin remaining, will have a lower detection rate and therefore narrower strip width, while strip width will also be narrower during the peak and late wet season.

To test for bias due to the object moving away from the observers, once the patrol staff had moved completely out of view, without informing the observers, he walked to either the left or the right for a particular distance, then back to the road. The perpendicular distance to the location where head and shoulders of the person re-appeared as first detected by the observers was measured and compared with the distance obtained when moving away from the road. Distances were measured with a Garmin GPSMap 60CSx using the 3D setting to limit the error to 1 to 2 m. Shorter distances in the Kakum forest, where satellite coverage was poor, were obtained with a tape measure.

After inspection of the underlying distribution of each data set, a Wilcoxon matched pairs test was done to test for differences between the moving-away and the moving-back data. When the difference was not significant, the two data sets were combined to estimate strip width. The resulting estimate of strip width should be considered an approximation of the maximum mean strip width for the six-month period covering the dry season. The mean for the six-month wet season was approached by comparing unburned areas with burned areas of the same vegetation type, wherever possible by comparing left and right at the same location for sites that were nearly completely burned (Murchison Falls and Mole). For Queen Elizabeth, of which approximately 35% was burned (estimate by management), we used measurements in burned areas to estimate dry-season mean strip width and those in unburned areas to estimate the mean for the wet season. We estimated the size of the area patrolled each year by multiplying the maximum mean strip width with the total distance patrolled, while correcting for the six-month period with lower visibility. The probability of detecting an elephant carcass on patrol was estimated by the proportion of the site covered by patrols for a particular year. Elephant carcasses found by patrol teams, for both natural and illegal or unknown deaths, except for those detected through the presence of vultures, were corrected for the area not covered by patrols in a particular year using the inverse of the detection probability. Carcasses corrected for detection probability were enumerated with those detected through information a priori (informers, general public, tourists and researchers) or through the presence of vultures in savannah sites (strip width between 1 and 3 km).



Using MIST data, we examined patrol distribution and the distributions of live and dead elephants for each year. Testing for randomness of patrols was not necessary, because the four sites selected are relatively small and were intensively patrolled on foot, applying a patrol strategy that aimed for total coverage per year (Murchison Falls and Mole) or per quarter (Kakum and Queen Elizabeth) whenever practically feasible. The spatial distribution of illegally killed elephants was compared with that of elephants that died of natural causes.

The corrected PIKE was then compared with the raw PIKE as estimated from the detailed carcass data from the MIKE database.

Carcass detection probability and cause of death

For relatively small sites with small elephant populations under sound management, the information provided by the protected area staff and MIKE carcass forms on the mode of detection of each carcass, combined with area coverage by patrols, gave approximations of the mean detection probability over a six-year period for each class of carcass, i.e. natural deaths, illegally killed elephants and unknown.

Comparing C/E results with PIKE

For each of the four sites, annual effort data were available in the form of distance covered on patrol (km) for the period from 2006 to 2011. The relationship between conventional patrol effort and poaching, which includes elephant poaching, follows a detection/deterrence curve, its shape determined by a number of factors such as size of the elephant population, size of the area, patrol effort and coverage, which includes the average size of patrol groups, poaching rate and visibility (Jachmann 1998, 2008). For example, in terms of visibility or habitat type, the curve peaks at much lower efforts in wide, open areas (grass-dominated savannah) compared with areas with low visibility, such as forest (Jachmann 2008). Initially, increasing patrol effort results in increased detection of poaching activities, resulting in a near linear to exponential ascending detection part of the curve that peaks at a short consolidation phase whereby effort and poaching levels are in equilibrium, followed by a descending deterrence phase, whereby the catch declines with increasing effort. As an example, the

relationship between serious poaching offences encountered per effective patrol man-day per month (catch) and increasing patrol effort (effective patrol man-days per month) for three forest sites in Ghana combined from 2005 to 2007 (Kakum, Ankasa and Bia Conservation Areas) is provided (Fig. 1). This relationship was best described by a third degree polynomial: Y = –1E – 0.7x³ + 0.0003x² – 0.0841x + 63.3020, P < 0.001 (Jachmann 2008). Prior to comparing C/E results with PIKE, we had to examine where the C/E data sets of the four sites were located on their respective detection/deterrence curves. We started by plotting patrol coverage and C/E elephants found killed illegally over time for each site. Then we inspected the relationships between C/E and PIKE and compared their trends. However, we should note that the C/E index is not independent from PIKE, because both have elephants killed illegally in the nominator. Therefore, results and discussion will merely focus on pitfalls and merits of both methods.

Comparing absolute mortality with PIKEFor the years 2006 to 2011, PIKE corrected for detection probability was compared with absolute elephant mortality (total mortality) for the four sites combined. Because few elephant counts had been done in Ghana, we used the results of the most recent ones (see section on Study areas above) to estimate total annual mortality for all sites combined.

Figure 1. Relationship between serious poaching offences encountered per effective patrol man-day/month and effective patrol man-days per month for three forest sites combined for 2005 to 2007 (Jachmann 2008).

200 400 600 800 1000 1200 1400 1600E�ective patrol man-days/month (2005–2007)

50

60

70

80

90

100

110

120

130

140

150

Serio

us p

oach

ing

o�en

ces

enco

unte

red/

e�ec

tive

patr

ol m

an-d

ay

0 50 100 150 200 250 300 350 400 450 500 550 600Distance (m) for back data for QENP

0

5

10

15

20

25

30

35

40

45

50

No.

of o

bser

vatio

ns


Jachmann

Results

Estimation of mean maximum strip widthQueen elizabeth national Park, uganda

The away and the back data followed a log-normal distribution (Fig. 2), but with significantly different means (Wilcoxon matched pairs test (T = 367.50, Z = 6.54, P = 0.0000), with the mean strip width for the back data being 25% lower than that for the away data, and the variance 24% lower (mean away = 136.53 ± 146.65, range 11–622; mean back = 102.44 ± 111.72, range 8–525). Due to bias in the away data, the back data were used to estimate the mean maximum strip width. The sample of 88 back measurements showed that 39% of the site was burned, which was close to the estimate provided by management. For burned areas, the mean strip width was 164.85 m, and for unburned areas it was 64.07 m. The mean maximum strip width was (6 x 329.70 + 6 x 128.14)/12 = 229 m.

Murchison Falls national Park, uganda

The away and the back data both followed the same approximate log-normal distribution (Fig. 3) with similar means (Wilcoxon matched pairs test (T = 1373.00, Z = 1.84, P = 0.0662), (mean away = 158.01 ± 85.92, range 31–403; mean back = 151.70 ± 87.36, range 18–404). The two sets of data were combined to estimate the mean maximum strip width. The sample of 176 strip width measurements showed that 93% of the site was burned. Corrected for the six-month wet season (33.4% lower visibility than dry season), mean maximum strip width was (6 x 310 + 6 x 206)/12 = 258 m.

Mole national Park, ghana

The away and the back data both followed an approximate log-normal distribution, but with significantly different means (Wilcoxon matched pairs test (T = 387.50, Z = 7.29, P = 0.0000), (mean away = 67.67 ± 30.55, range 22–241; mean back = 52.95 ± 27.74, range 18–216). The back data were 21.8% lower than the away data and were used to estimate the mean maximum strip width (Fig. 4). Our sample of 100 back measurements showed that 96% of the site was burned. Although based on a small sample size (n = 4), visibility in unburned areas (wet season) was roughly 40% lower than in burned areas (dry season). Corrected for the six-month wet season period with lower visibility, the mean maximum strip width was (6 x 106 + 6 x 42)/12 = 74 m.

Figure 2. Frequency diagram and distribution of back data for Queen Elizabeth NP (QENP).

200 400 600 800 1000 1200 1400 1600E�ective patrol man-days/month (2005–2007)

50

60

70

80

90

100

110

120

130

140

150

Serio

us p

oach

ing

o�en

ces

enco

unte

red/

e�ec

tive

patr

ol m

an-d

ay

0 50 100 150 200 250 300 350 400 450 500 550 600Distance (m) for back data for QENP

0

5

10

15

20

25

30

35

40

45

50

No.

of o

bser

vatio

ns

0 50 100 150 200 250 300 350 400 450Distance (m) for back and away data for MFNP

0

5

10

15

20

25

30

35N

o. o

f obs

erva

tions

0 20 40 60 80 100 120 140 160 180 200 220 240Distance (m) for back data for Mole NP

0

5

10

15

20

25

30

35

40

45

50

No.

of o

bser

vatio

ns

Figure 3. Frequency diagram and distribution of away data and back data combined for Murchison Falls National Park (MFNP), North Bank.

0 50 100 150 200 250 300 350 400 450Distance (m) for back and away data for MFNP

0

5

10

15

20

25

30

35

No.

of o

bser

vatio

ns

0 20 40 60 80 100 120 140 160 180 200 220 240Distance (m) for back data for Mole NP

0

5

10

15

20

25

30

35

40

45

50

No.

of o

bser

vatio

ns

Figure 4. Frequency diagram and distribution of back data for Mole NP.



trails, management enforces strict rules that elephant trails or other existing pathways are not to be used during patrols. This implies that even at this patrol density, duplicate routes may be rare. Because all four sites were completely covered at least annually, there was no need to perform tests of randomness of patrol routes.

Estimation of detection probability

Distances and areas patrolled, detection probabilities and correction factors for each of the four sites are provided in Tables 1 to 4. In Murchison Falls NP, numbers of patrols using a GPS, hence patrol data entered into the MIST system, varied between roughly

kakuM conservation area, ghana

The away and the back data both followed an approximate log-normal distribution, but with significantly different means (Wilcoxon matched pairs test (T = 338.50, Z = 2.21, P = 0.0273), (mean away = 20.68 ± 8.23, range 3–37; mean back = 17.56 ± 7.96, range 3–45). The back data (Fig. 5) were 15.1% lower than the away data and were used to estimate the mean maximum strip width (2 x 17.56 m) = 35 m.

Summary of site mean maximum strip widthsApproximate log-normal distributions of strip width for the four sites show the pronounced differences in visibility profiles, thus detection probabilities (Fig. 6). Kakum CA consists predominantly of moist secondary forest with pockets of primary forest on mildly undulating terrain; it had a mean maximum strip width of 35 m. Mole NP consists primarily of woodland savannah on mildly undulating terrain; it had a mean strip width of 74 m. Queen Elizabeth NP was covered by a mosaic of different vegetation types but was dominated by open grassland on mildly undulating terrain; mean strip width was 229 m. Murchison Falls NP consisted primarily of wide open grassland with Borassus aethiopum and Acacia woodlands on mildly undulating terrain but with sharp ridges; mean strip width was 258 m.

Patrol coveragePark management in Queen Elizabeth NP aims to have patrols cover the entire site on a quarterly basis, without duplicating a single patrol path. Although management of Murchison Falls NP aims for patrols to cover the entire park annually but with emphasis on the North Bank, only 30–40% of all patrol data were entered into MIST. There was no information on duplicate patrol routes. Park management in Mole NP aims to have patrols cover the entire site annually, with all patrols using a GPS and all patrol data incorporated in MIST. As the site is savannah with easy access throughout, according to management, duplicate patrol routes are rare. Management of Kakum aims to have patrols cover the entire site every quarter. Although Kakum is a forest site, where patrols tend to use existing elephant

0 5 10 15 20 25 30 35 40 45 50Distance (m) for back data for KCA

0

2

4

6

8

10

12

14

16

18

No.

of o

obse

rvat

ions

0 100 200 300 400 5000

20

40

60

80

100

120

140

160

180

200

220

240

No.

of o

bser

vatio

ns

Murchison Falls National Park

Queen Elizabeth National Park

Mole National Park

Kakum Conservation Area

Figure 5. Frequency diagram and distribution of back data for Kakum Conservation Area (KCA).

0 5 10 15 20 25 30 35 40 45 50Distance (m) for back data for KCA

0

2

4

6

8

10

12

14

16

18

No.

of o

obse

rvat

ions

0 100 200 300 400 5000

20

40

60

80

100

120

140

160

180

200

220

240

No.

of o

bser

vatio

ns

Murchison Falls National Park

Queen Elizabeth National Park

Mole National Park

Kakum Conservation Area

Figure 6. Log-normal distribution of strip width data for the four sites.


Jachmann

30% and 40%. Moreover, before 2011 few patrols were carried out on the South Bank of the park. Therefore, we may conclude that the North Bank, which is also the main tourist area, was always entirely covered. Thus detection probabilities for the years 2006 to 2011 (Table 2) are most likely much higher, approaching 1 in most years, which implies that the carcass data did not need to be corrected for detection probability. However, due to limited incorporation of patrol data

into the MIST system, results for Murchison Falls should be interpreted with caution.

For each of the four sites we calculated patrol density (coverage) in terms of distance covered on patrol (km) per square kilometre of site. By plotting detection probability against patrol density, we estimated the minimum patrol density for which the probability of detecting an elephant carcass approached 1.00 (Tables 1 to 4). Using grassland cover in each of the four sites

Table 1. Distance and area patrolled, and detection probability of carcasses in Queen Elizabeth NP, UgandaYear Distance patrolled (km) Area patrolled (km²) Detection probability Correction factor2006 7,245 1,659 0.72 1.392007 5,531 1,267 0.55 1.822008 7,243 1,659 0.72 1.392009 7,809 1,788 0.78 1.282010 10,439 2,390 1.00 1.002011 10,339 2,368 1.00 1.00

The area site is 2,294 km², strip width 0.229 km for all years.

Table 2. Distance and area patrolled, and detection probability of carcasses in Murchison Falls NP, UgandaYear Distance patrolled (km) Area patrolled (km²) Detection probability Correction factor2006 6,333 1,634 1.00 1.002007 4,228 1,091 0.73 1.372008 6,212 1,603 1.00 1.002009 4,744 1,224 0.82 1.222010 2,989 771 0.51 1.962011 4,706 1,214 0.81 1.24

Area site is 1.500 km², strip width 0.258 km for all years.

Table 3. Distance and area patrolled, and detection probability of carcasses in Mole NP, GhanaYear Distance patrolled (km) Area patrolled (km²) Detection probability Correction factor2006 58,866 4,356 0.97 1.032007 48,314 3,575 0.79 1.272008 65,832 4,872 1.00 1.002009 81,437 6,026 1.00 1.002010 107,777 7,976 1.00 1.002011 122,528 9,067 1.00 1.00

Area site is 4,505 km², strip width 0.074 km for all years.

Table 4. Distance and area patrolled, and detection probability of carcasses in Kakum CA, GhanaYear Distance patrolled (km) Area patrolled (km²) Detection probability Correction factor2006 8,693 304 0.83 1.212007 9,206 322 0.88 1.142008 12,765 447 1.00 1.002009 12,170 426 1.00 1.002010 14,696 514 1.00 1.002011 10,857 380 1.00 1.00

Area site is 366 km², strip width 0.035 km for all years.



evenly spread over the site. Although inspection of field data on site showed that more or less the same applied to Murchison Falls, most carcass data were missing in the MIST database.

Corrected PIKE versus MIKE PIKEFor Queen Elizabeth, Murchison Falls and Mole NPs, PIKE obtained from verified field data and corrected for detection probability closely followed PIKE calculated from the detailed carcass sheets obtained from MIKE/CCU (Table 10, Figs. 8–10).

Due to the small numbers of carcasses and minor inaccuracies in reporting from the field for Kakum Conservation Area, PIKE obtained from verified field data and corrected for detection probability deviated from PIKE calculated from the detailed carcass sheets (Fig. 11).

Cause of death and detection probabilityThe samples were too small to compare detection probabilities for the three different categories of carcasses for the four different sites. However, for the four sites combined, the mean detection probability for elephants that died of natural causes (0.887 +/- 0.125

Table 5. Visibility (percentage of grassland cover) and minimum patrol density (patrol distance (km/km²) for a carcass detection probability of 1.00, for four sites

Site Grassland (%)

Rank Patrol density (km/km²)

Murchison Falls NP 60 4 4.0Queen Elizabeth NP 40 6 4.4Mole NP 10 9 14.5Kakum CA 0 10 29.5

0 10 20 30 40 50 60 70

Visibility (% grassland cover)

0

5

10

15

20

25

30

35

Patr

ol d

ensi

ty (k

m/k

m2 )

2005 2006 2007 2008 2009 2010 2011 2012Year

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

PIK

E

PIKE from detailed carcass sheets (MIKE) PIKE from veri�ed corrected �eld data

Figure 7. Patrol density at detection probability 1.00 against visibility (percentage of grass cover) for four sites.

as a measure of visibility, with 0 grassland cover ranked as 10 and 100% grassland cover ranked as 0, we plotted patrol density at detection probability 1.00 against visibility (Table 5, Fig. 7).

The relationship between patrol density (coverage) at which the detection probability approaches 1.00, and visibility in terms of percentage of grass cover was negative exponential (y = 1.384 + e(3.166 – 0.047x); R = 0.97, variance explained = 94.356%), most likely running asymptotic at roughly 60% grass cover (Fig. 7). Although this relationship is based on the data of four sites only, it suggests that independent of grass cover, thus visibility, a minimum patrol density of roughly 4 patrol km/km² of site is required (Fig. 7).

Mode of detecting carcassesFor each carcass found in the four sites and their immediate surroundings, the mode of detection was obtained from the MIKE carcass sheets, complemented with information provided by management and patrol staff (Tables 6 to 9). For natural, illegal and unknown deaths, carcasses were divided into three categories: those found with prior information through the intelligence network, tourists, general public and researchers; those found on regular foot patrols; and those found through the presence of vultures. Because the presence of vultures increased strip width to anywhere between 1 and 3 km, these carcasses did not require correction for detection probability. With the exception of Murchison Falls, carcasses found during regular foot patrols were corrected for detection probability. Because the MIST data on patrol statistics for Murchison Falls were not reliable, carcass data could not be corrected for detection probability. However, as concluded in a previous section, detection probability of all classes of carcasses must have been close to 1.00, and for Murchison Falls we will merely compare PIKE derived from the detailed carcass data (MIKE) with the raw but verified field data.

Spatial distribution of carcassesFor Kakum CA and Mole NP in Ghana, numbers of carcasses were too low to examine spatial distribution. Elephant carcasses found in Queen Elizabeth between 2002 and 2011 were evenly distributed over the park, much in line with elephant distribution, while the cause of death was of no influence on the spatial distribution of carcasses, with natural, illegal and unknown deaths


Jachmann

Table 6. Carcass numbers by mode of detection for Queen Elizabeth NP, with carcasses corrected for detection probability in brackets

Carcass numbers per yearMode of detection 2006 2007 2008 2009 2010 2011Legal 0 0 0 0 0 1Natural Prior information 1 0 2 1 1 0 Regular patrol 2 (2.8) 0 2 (2.8) 1 (1.3) 0 0 Vultures 0 0 0 0 1 0Illegal Prior information 0 0 0 1 1 7 Regular patrol 2 (2.8) 2 (3.6) 0 2 (2.6) 3 12 Vultures 0 0 0 0 0 1Unknown Prior information 1 4 1 1 1 0 Regular patrol 0 2 (3.6) 1 (1.4) 2 (2.6) 4 4 Vultures 0 0 0 0 0 0Total 6 (7.6) 8 (11.2) 6 (7.2) 8 (9.5) 11 25

Table 7. Carcass numbers by mode of detection for Murchison Falls NP, 2006–2011Carcass numbers per year

Mode of detection 2006 2007 2008 2009 2010 2011Legal 0 0 0 1 0 0Natural Prior information 0 0 0 0 1 0 Regular patrol 0 1 1 0 0 4 Vultures 0 0 0 0 0 0Illegal Prior information 0 0 0 1 0 7 Regular patrol 1 0 2 2 3 9 Vultures 0 0 0 0 0 0Unknown Prior information 0 0 0 2 1 0 Regular patrol 1 0 1 0 2 4 Vultures 0 0 0 0 0 0Total 2 1 4 6 7 24

Table 8. Carcass numbers by mode of detection for Mole NP, with carcasses corrected for detection probability in brackets

Carcass numbers per yearMode of detection 2006 2007 2008 2009 2010 2011Legal 0 0 0 0 0 0Natural Prior information 0 0 0 0 0 0 Regular patrol 0 0 0 0 0 0 Vultures 0 0 0 0 0 0Illegal Prior information 0 0 0 0 0 0 Regular patrol 2 (2.1) 2 (2.5) 3 0 2 0 Vultures 0 1 0 0 0 0Unknown Prior information 0 1 0 0 0 1 Regular patrol 0 1 (1.3) 1 0 0 0 Vultures 0 0 0 0 0 1Total 2 (2.1) 5 (5.8) 4 0 2 2



0 10 20 30 40 50 60 70

Visibility (% grassland cover)

0

5

10

15

20

25

30

35

Patr

ol d

ensi

ty (k

m/k

m2 )

2005 2006 2007 2008 2009 2010 2011 2012Year

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

PIK

E


Figure 8. Comparison of proporation of illegally killed elephants (PIKE) from detailed carcass sheets (MIKE/CCU) with verified field data, corrected for detection probability, for Queen Elizabeth NP.

2005 2006 2007 2008 2009 2010 2011 2012Year

-0.10.00.10.20.30.40.50.60.70.80.91.0

PIKE

2005 2006 2007 2008 2009 2010 2011 2012

Year

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

PIKE


PIKE from detailed carcass sheets (MIKE)

PIKE from veri�ed corrected �eld data

Figure 9. Comparison of proporation of illegally killed elephants (PIKE) from detailed carcass sheets (MIKE/CCU) with verified field data, for Murchison Falls NP.

2005 2006 2007 2008 2009 2010 2011 2012Year

-0.10.00.10.20.30.40.50.60.70.80.91.0

PIKE

2005 2006 2007 2008 2009 2010 2011 2012

Year

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

PIKE


PIKE from detailed carcass sheets (MIKE)

PIKE from veri�ed corrected �eld data

Figure 10. Comparison of proporation of illegally killed elephants (PIKE) from detailed carcass sheets (MIKE/CCU) with verified field data for Mole NP.

2005 2006 2007 2008 2009 2010 2011 2012Year

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2 P

IKE


2005 2006 2007 2008 2009 2010 2011 2012Year

0

5

10

15

20

25

30

35

40

45

Patr

ol c

over

age

(km

/km

2 of s

ite)

Queen Elizabeth Murchison Falls Mole Kakum

Figure 11. Comparison of proporation of illegally killed elephants (PIKE) from detailed carcass sheets (MIKE/CCU) with verified field data, corrected for detection probability, for Kakum Conservation Area.

Table 9. Carcass numbers by mode of detection for Kakum CA, with carcasses corrected for detection probability in brackets, 2006–2011

Carcasses numbersMode of detection 2006 2007 2008 2009 2010 2011Legal 0 0 0 0 0 0Natural Prior information 0 0 0 0 0 0 Regular patrol 0 0 0 0 1 0 Vultures 0 0 0 0 0 0Illegal Prior information 0 0 1 0 1 0 Regular patrol 1 (1.2) 0 0 1 1 1 Vultures 0 0 0 0 0 0Unknown Prior information 0 1 0 0 0 0 Regular patrol 0 0 0 0 0 0 Vultures 0 0 0 0 0 0Total 1 (1.2) 1 1 1 3 1


Jachmann

Table 11. Detection probability by cause of death for four sites Cause of death and detection probability

Site Natural Illegal UnknownQueen Elizabeth National Park 0.88 0.94 0.92Murchison Falls National Park 0.85 0.86 0.84Mole National Park – 0.99 0.98Kakum Conservation Area 1.00 0.97 1.00Combined +/– SD 0.887 +/– 0.125 0.887 +/– 0.151 0.906 +/– 0.157

– = no data

Table 10. Comparing PIKE from detailed carcass sheets with verified field data corrected for detection probability

PIKE comparison 2006 2007 2008 2009 2010 2011Queen Elizabeth NP Carcass sheets 0.33 0.25 0 0.38 0.36 0.74 Field data 0.37 0.32 0 0.38 0.36 0.80Murchison Falls NP Carcass sheets 0.50 0 0.50 0.40 0.29 0.88 Field data 0.50 0 0.50 0.50 0.43 0.67Mole NP Carcass sheets 1.00 0.80 0.75 0 1.00 – Field data 1.00 0.60 0.75 0 1.00 0.50Kakum CA Carcass sheets 0 0 1.00 1.00 0 – Field data 1.00 0 1.00 1.00 0.67 1.00

– = no data

(SD)) was identical to that for those killed illegally (0.887 +/- 0.151) (Table 11). Detection probability for carcasses in the ‘unknown’ category was slightly higher (0.906 +/- 0.157) than in the other two categories (Table 11), and significantly different from those that died of natural causes (sign test, Z = 2.846, P = 0.004), but not from those killed illegally (Z = 0.267, P = 0.789). This difference stems from the fact that 37% of the carcasses in the unknown category were found through information systems (intelligence networks, tourists, researchers, or general public), which compares with 29% of those that died of natural causes. Because the majority of carcasses in the unknown category were old, the cause of death was difficult to determine, but there had been ample time for these carcasses to be detected through another information system.

Comparing elephants found killed illegally/patrol km (C/E) with PIKE

For this entire section, elephants found killed illegally by patrols includes those found through information channels, while PIKE was calculated from the raw carcass sheets, validated in the field, and corrected for detection probability. Elephant carcasses found

by patrols using prior information only involved a few years, mostly concerning 2011 for the two sites in Uganda and two carcasses in Kakum, while analyses omitting these carcasses gave identical results. For the period 2006 to 2011, the size of patrol groups was between about four and five for each of the four sites. Throughout this section, patrol density and patrol coverage are used alternately, but both apply to patrol km/km² of site. Furthermore, as discussed in a previous section, the C/E index and PIKE are not independent, because both have elephants killed illegally in the nominator. Taking into account differences in habitat types, hence visibility, we were able to compare patrol density (coverage) in terms of patrol km per km² of site (Fig. 12).

Patrol density required for a carcass detection probability of 1 will be lower than that required to deter poachers from entering a conservation area. Information on patrol density merely provided an indication on where approximately the C/E results were located on their respective detection/deterrence curves (see Fig. 1). With C/E results located around the peak and descending part of the curve, comparing with PIKE may give unreliable results. Moreover, PIKE, given stable natural and legal mortality, follows



2005 2006 2007 2008 2009 2010 2011 2012Year

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

PIK

E


2005 2006 2007 2008 2009 2010 2011 2012Year

0

5

10

15

20

25

30

35

40

45

Patr

ol c

over

age

(km

/km

2 of s

ite)

Queen Elizabeth Murchison Falls Mole Kakum

Figure 12. Patrol density (coverage) in patrol km/km² of site for Queen Elizabeth and Murchison Falls NPs in Uganda, and Kakum Conservation Area and Mole NP in Ghana, 2006–2011.

times the patrol density of Queen Elizabeth and hardly any elephant poaching, with C/E most likely in the deterrence part of the curve. On the other hand, effort in Kakum, being a forest area with low visibility but high patrol density, may have been along the top of the curve, with parts in the ascending detection phase. In Queen Elizabeth and Murchison Falls, effort was most likely still in the mostly linear ascending part of the curve. With only 30–40% of patrol data entered in MIST, comparing C/E results with PIKE for Murchison Falls may give unreliable results, but those for Queen Elizabeth may compare well.

For Queen Elizabeth NP, the C/E trend in terms of elephants found killed illegally/patrol km roughly follows that of PIKE (Fig. 13). Because the data sets were small, the underlying distributions of C/E and PIKE were not significantly different from normal (Kolmogorov-Smirnov test; C/E, d = 0.432, P < 0.2; PIKE, d = 0.272, P > 0.2), and the linear relationship was significant at P = 0.012 (Fig. 14). As an additional test, the Spearman rank correlation was also significant at P < 0.05 (r = 0.943). For Queen Elizabeth, C/E analyses provided support for the use of PIKE to monitor elephants killed illegally, albeit at relatively low levels of patrol density and at relatively low levels of elephant poaching.

For Murchison Falls NP, albeit with strong divergence of individual data points, both the C/E index and PIKE showed an upward trend in elephant poaching from 2007 onwards (Fig. 15). Due to this divergence in data points, the relationship between the C/E index and PIKE was not significant.

For Mole NP, with the exception of 2006 and 2007, the trend of the C/E index roughly followed that of

an asymptotic function, gradually slowing down at higher levels of poaching, and levelling off when it approaches 1 (Jachmann 2012). Therefore, even when C/E results are located on the mostly linear ascending part of the curve, when comparing these results with PIKE values > 0.7 (Jachmann 2012), they may not be reliable. During the study period, patrol density (coverage) in Queen Elizabeth slowly increased but remained under 5 km/km² of site, whereas coverage in Murchison Falls sharply dropped from more than 40 km/km² in 2006 to less than 5 km/km² in 2007, and remained well below 5 km/km² up to 2011 (Fig. 12). Due to anomalies in data collection and incorporation into the MIST system, as outlined above, the results for Murchison Falls National Park should be interpreted with caution. In Mole, patrol density steeply increased from slightly below 15 km/km² in 2006 to about 27 km/km² of site in 2011, while Kakum showed a similar steep increase from about 14 km/km² in 2006 to about 40 km/km² of site in 2010, and then sharply dropped to about 29 km/km² in 2011 (Fig. 12).

With mean maximum strip width used as a measure of visibility, Mole, being dominated by woodland savannah, had more than twice the visibility of the Kakum forest environment (0.074/0.035 = 2.11), and therefore theoretically Kakum required more than twice the patrol coverage of Mole for a similar deterrence effect. Murchison Falls and Queen Elizabeth, dominated by grassland savannah, had respectively 3.49 and 3.10 times better visibility than Mole and therefore theoretically required only one-third the patrol coverage of Mole for a similar deterrence effect. However, in 2011, Mole had more than five

2005 2006 2007 2008 2009 2010 2011 2012Year

-0.0002

0.0002

0.0006

0.0010

0.0014

0.0018

0.0022

Elep

hant

s fo

und

kille

d ill

egal

ly/p

atro

l km

(C/E

)

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

PIKE

C/E PIKE

-0.0002 0.0002 0.0006 0.0010 0.0014 0.0018 0.0022

Elephants found killed illegally/patrol km (C/E)

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

PIK

E

Figure 13. Comparison of the trend in C/E with that of proportion of illegally killed elephants (PIKE) for Queen Elizabeth NP, 2006–2011.


Jachmann

PIKE (Fig. 16). Consistently high and increasing patrol coverage deterred most elephant poaching, resulting in small numbers of carcasses with C/E data mainly distributed over the peak and descending deterrence parts of the curve, causing high variability (Fig. 16). The relationship between the C/E index (elephants found killed illegally/patrol km) and PIKE was not significant but otherwise provided a textbook example of a detection/deterrence curve (Fig. 17).

For the Kakum Conservation Area, with the exception of 2010, the trend in PIKE closely followed that of the C/E index (Fig. 18). However, the relationship between the C/E index and PIKE was not significant. Throughout the study period, patrol coverage in Kakum was sufficiently high to deter most elephant poaching, which resulted in few elephants killed illegally, a PIKE data set containing mostly ones and zeros, and C/E data distributed over the peak and ascending detection parts of the detection/deterrence curve. In general terms, the trend in C/E for Kakum supports the use of PIKE for monitoring elephant poaching.

For Queen Elizabeth, Mole and Kakum combined, with the exception of 2007 the trend in PIKE was similar to that of the C/E index (Fig. 19). The underlying distributions of C/E and PIKE were not significantly different from normal (Kolmogorov-Smirnov test: C/E, d = 0.223, P > 0.2; PIKE, d = 0.164, P > 0.2), but due to the divergence in data points for 2007, the linear relationship was barely significant at the 10% level (Fig. 20; P = 0.100). If 2007 is omitted, a highly significant linear relationship emerges (P = 0.0137).

In summary, because the shape of the detection/deterrence curve depends on a number of factors that may vary by site, C/E analysis remains a complicated method to monitor elephant poaching using aggregated data from a large number of sites, frequently with pronounced differences in habitat type, poaching, law enforcement and other relevant factors. When C/E is compared with PIKE, the latter has its limitations due to its asymptotic function, but in spite of a number of potential biases in the method, it remains superior to C/E analysis to monitor poaching trends using aggregated data from sites that vary in at least a few but often most of the factors discussed above.

Comparing absolute mortality with PIKEWith the exception of 2006, in general terms, the trend

2005 2006 2007 2008 2009 2010 2011 2012Year

-0.0002

0.0002

0.0006

0.0010

0.0014

0.0018

0.0022

Elep

hant

s fo

und

kille

d ill

egal

ly/p

atro

l km

(C/E

)

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

PIKE

C/E PIKE

-0.0002 0.0002 0.0006 0.0010 0.0014 0.0018 0.0022


-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

PIK

E

Figure 14. The relationship between proportion of illegally killed elephants (PIKE) and C/E for Queen Elizabeth NP, 2006–2011 (broken lines represent the 95% confidence limits).

Figure 15. Comparison of the trend in C/E with that of proportion of illegaly killed elephants (PIKE) for Murchison Falls NP, 2006–2011.

2005 2006 2007 2008 2009 2010 2011 2012

Year

–0.0005

0.0000

0.0005

0.0010

0.0015

0.0020

0.0025

0.0030

0.0035

0.0040

Elep

hant

s fo

und

kille

d ill

egal

ly/p

atro

l km

(C/E

)

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

PIK

E

C/E PIKE

2005 2006 2007 2008 2009 2010 2011 2012

Year

–0.00001

0.00000

0.00001

0.00002

0.00003

0.00004

0.00005

0.00006

0.00007

Elep

hant

s fo

und

kille

d ill

egal

ly/p

atro

l km

(C/E

)

–0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

PIK

E

C/E PIKE

2005 2006 2007 2008 2009 2010 2011 2012

Year

–0.0005

0.0000

0.0005

0.0010

0.0015

0.0020

0.0025

0.0030

0.0035

0.0040

Elep

hant

s fo

und

kille

d ill

egal

ly/p

atro

l km

(C/E

)

-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

PIK

E

C/E PIKE

2005 2006 2007 2008 2009 2010 2011 2012

Year

–0.00001

0.00000

0.00001

0.00002

0.00003

0.00004

0.00005

0.00006

0.00007

Elep

hant

s fo

und

kille

d ill

egal

ly/p

atro

l km

(C/E

)

–0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

PIK

E

C/E PIKE

Figure 16. Comparison of the trend in C/E with that of proportion of illegaly killed elephants (PIKE) for Mole NP, 2006–2011.



-0.00001 0.00001 0.00003 0.00005 0.00007Elephants found killed illegally/patrol km (C/E)

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

PIK

E

Figure 17. The relationship between proportion of illegally killed elephants (PIKE) and C/E (elephants found killed illegally/patrol km) for Mole NP, 2006–2011.

2005 2006 2007 2008 2009 2010 2011 2012Year

–0.0002

0.0002

0.0006

0.0010

0.0014

Elep

hant

s fo

und

kille

d ill

egal

ly/

patr

ol k

m (C

/E)

–0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2PI

KE

C/E(L) PIKE(R)

Figure 18. Comparison of the trend in C/E with that of proportion of illegally killed elephants (PIKE) for Kakum Conservation Area, 2006–2011.

2005 2006 2007 2008 2009 2010 2011 2012Year

0.30

0.35

0.40

0.45

0.50

0.55

0.60

0.65

0.70

0.75

PIKE

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Abs

olut

e m

orta

lity

(%)

PIKE corrected for detection probability Absolute mortality corrected for detection probability

Figure 21. Comparison of the trend in PIKE, corrected for detection probability, with that in absolute mortality for the four sites combined, 2006–2011.

Figure 20. The relationship between proportion of illegally killed elephants (PIKE) and C/E for Queen Elizabeth and Mole NPs and Kakum CA combined; 2006–2011 (broken lines represent the 95% confidence limits).

0.00002 0.00006 0.00010 0.00014


0.25

0.35

0.45

0.55

0.65

0.75

PIKE

0.00018

Figure 19. Comparison of the trend in C/E with that of proportion of illegally killed elephants (PIKE) for Queen Elizabeth and Mole NPs, and Kakum Conservation Area combined, 2006–2011.

2005 2006 2007 2008 2009 2010 2011 2012

Year

0.00002

0.00004

0.00006

0.00008

0.00010

0.00012

0.00014

0.00016

Elep

hant

s fo

und

kille

d ill

egal

ly/

patr

ol k

m (C

/E)

0.25

0.35

0.45

0.55

0.65

0.75

PIK

E

C/E PIKE

in PIKE follows that of absolute mortality in the four sites combined (Fig. 21). With more up-to-date and a longer series of survey results, the trend in absolute mortality could be refined and may follow that of PIKE more closely. Thus, PIKE may also be used to monitor absolute mortality.

DiscussionThroughout the report, the mean strip width was dubbed the mean maximum strip width for a good reason. The strip width measured applied to the detection of relatively fresh carcasses with a height of roughly 1.3 m, but not to old carcasses in various stages of decay. For older carcasses, the strip width will be narrower. We considered incorporating a correction factor for older carcasses in the estimation of strip width but refrained from doing so simply because without sufficient research combined with a small data set, we would merely further weaken the results. Moreover, although management at three out of four sites insisted that patrols never used duplicate routes, upon inspecting patrol coverage we were inclined to doubt these statements. The higher the frequency of duplicate patrol routes, the lower the probability of


Jachmann

detecting a carcass. In other words, our estimates of strip width may be positively biased—that is, they may be too wide—and so may be our estimates for patrol density (coverage).

Both of these biases may have resulted in overestimation of detection probability and underestimation of carcass densities. However, over the past six years, in three of the sites visited, law-enforcement effort in terms of patrol distance covered per square kilometre of site increased by 69% in Kakum (2006 to 2010), 108% in Mole, and 30% in Queen Elizabeth. Therefore, it is highly likely that even with a narrower strip and correction for duplicate patrol routes, during the end of the study period, the probability of detecting an elephant carcass may have approached 1 in each of these sites. The same may have applied to Murchison Falls, with only 30–40% of all patrols entered into the MIST system. This, however, does not account for under-sampling of juvenile deaths, especially the youngest ones (< 1 year). All of the above biases may have led to underestimates of carcass densities.

PIKE may also be biased because of differential detection probabilities for different causes of mortality and because of background variation in elephant mortality (Burn et al. 2011). Although our study did not find any evidence with regard to differential detection probabilities for different causes of mortality, the sample was small and pertained to four sites that were well managed and intensively covered by patrols. Within the MIKE sample of 80 sites there are undoubtedly many sites that are not well managed or properly patrolled, where detection probabilities may vary by cause of death. Background variation in elephant mortality may be caused by adverse environmental conditions, such as prolonged drought (Burn et al. 2011). For the period and sites covered in this pilot, we did not find any adverse environmental conditions. Another source of bias may be hidden in the spatial distribution of elephant carcasses, especially where the distribution of elephants that died of natural causes differs from those killed illegally. If a statistically significant difference exists between the spatial distribution of elephants killed illegally and those that died of natural causes, detection probabilities may differ, depending on patrol density. Albeit based on limited data, this pilot study did not find any evidence for differential spatial distribution of carcasses by cause of death. Moreover, for the four sites combined, detection

probabilities for elephants that died of natural causes and those killed illegally were exactly the same (Table 11). These biases, however, may balance when sample size is large enough.

An interesting observation on detection probability of carcasses in the savannah is that out of a total of 123 carcasses, only 4 were detected through the presence of vultures and none through the presence of other scavengers. This information was derived not only from carcass sheets and patrol forms but mostly through interviews with patrol staff who were present when a carcass was found. Moreover, in the forest environment of Kakum, 3 out of 8 carcasses were found through information channels (37.5%), while in the three savannah sites, 36 carcasses were found through prior information (29.3%). Although the sample is relatively small, it just may shed some doubt on the generally accepted idea that in the savannah most elephant carcasses are detected through the presence of vultures or other scavengers, and that in the forest most carcasses are found through information channels (intelligence, tourists, researchers and information provided by the general public).

It may be a few more years before sound effort data come available for more than the above four sites. However, data gaps in MIST need to be filled, sites should aim to incorporate all patrol data into MIST, all patrols need to carry a GPS, data should frequently be backed up, and problems with individual systems need to be resolved. To accomplish this, however, requires resources well beyond the current capacity of MIKE, implying that the responsible authorities in the countries involved will have to cover at least part of these shortfalls. In Ghana, data gaps and shortcomings in MIST were complemented with information from the manual system that was set up in 2004 (Jachmann 2004). Without the manual system in place, incomplete MIST data would have been insufficient to perform the analyses detailed above. It would be prudent to set up a similar system for all MIKE sites. Although C/E analysis provided support for the use of PIKE at the site level, it also showed that using patrol effort in C/E analysis has its intrinsic problems, primarily related to the polynomial function of the detection/deterrence curve. The curve shape depends on a number of factors that may vary by site.

Thus, when using effort in C/E instead of PIKE to monitor elephant poaching throughout their range in a wide variety of habitat types and other variables, the key is to find a generalized model that fits all different



offered both their precious time and their assistance to facilitate data collection. Special thanks to staff of the Uganda Wildlife Authority and the Ghana Wildlife Division for granting permission to visit the sites. We are most grateful to Ken Burnham and one unknown reviewer for providing valuable comments on an earlier draft.

ReferencesBouché, P. 2007. Northern Ghana elephant survey.

Pachyderm 42:58–69.Burn, R.W., Underwood, F.M. and Blanc J. 2011. Global

trends and factors associated with the illegal killing of elephants: A hierarchical Bayesian analysis of carcass encounter data. PLoS ONE 69: e24165.

Danquah, E.K.A. 2004. CITES–MIKE–Kakum National Park retrospective elephant survey 2004. A Rocha Ghana, Accra. Available at: http://www.cites.org/common/prog/mike/sub_regNN_Africa/ghana_2 004_survey.pdf.

Jachmann, H. 1998. Monitoring illegal wildlife use and law enforcement in African savannah rangelands. Lusaka: Environmental Council of Zambia.

Jachmann, H. 2004. Monitoring law enforcement for adaptive management: A manual. Wildlife Division, Accra. 124 p.

Jachmann, H. 2008. Illegal wildlife use and protected area management in Ghana. Biological Conservation 141:1906–1918.

Jachmann, H. 2012. Pilot study to validate PIKE-based inferences at the site level. CITES/MIKE technical report. 56 p.

Kahindi, O., Wittemeyer, G., King, J., Ihwagi, F., Omondi, P. and Douglas-Hamilton, I. 2009. Illegal killing of elephants across diverse land uses in Laikipia–Samburu, Kenya. African Journal of Ecology 48:972–983.

Plumptre, A., Kujirakwinja, D., Moyer, D., Driciru, M. and Rwetsiba, A. 2010. Greater Virunga landscape: large mammal surveys, 2010. Kampala, Uganda: Wildlife Conservation Society. Available at: http://www.elephantdatabase.org/population_submission/64.

Rwetsiba, A. and Wanyama, F. 2010. Aerial surveys of medium large mammals in Kidepo Valley and Murchison Falls Conservation Areas. Monitoring and Research Unit, Uganda Wildlife Authority, Kampala. Available at: http://www.elephantdatabase.org/population_submissions/63.

shapes of the curve. Elsewhere it has been suggested that to account for detection versus deterrence, a dynamic model is required that uses data of individual patrols rather than the site aggregates by year (Burn et al. 2011). Although this is undoubtedly true, practice shows it has been a major feat to obtain sound by-year aggregates from only a handful of sites. In practice, obtaining detailed data by patrol for 80 different sites will prove to be next to impossible. We may as well conclude that for MIKE’s purposes—that is, to monitor elephant poaching at sub-regional, regional and continental levels, due to the highly variable nature of the detection/deterrence curve, C/E analysis is too complicated and therefore not a practical approach to achieving this objective. PIKE data, on the other hand, can be easily pooled, and the pooling is likely to erode some of its imperfections—in other words, balance out some of its biases. As concluded earlier, PIKE is superior to C/E analysis for a system like MIKE. Moreover, a three-year study in the Laikipia–Samburu area of Kenya showed that PIKE offered a useful metric for comparing levels of illegal offtake temporarily and spatially, while its trends were relatively robust to systematic differences in methodology and spatial differences in data collection (Kahindi et al. 2009). In addition, PIKE may even prove to be a useful measure to monitor absolute elephant mortality.

In the meantime, based on the information collected in four relatively small and well-managed sites, PIKE, we may conclude, may prove to be a promising measure for monitoring elephant poaching at different spatial and temporal levels. However, the exercise described in this paper should be repeated for a sample of sites at the other end of the quality range—that is, some poorly managed sites with high illegal offtake, a clumped elephant distribution, and a patrolling density that is spatially irregular and low. These sites, however, should not be so poorly managed that information cannot be retrieved.

Acknowledgements

First, I would like to express my gratitude to the sub-regional support officer for eastern Africa, Edison Nuwamanya, and the regional director of the Wildlife Division in Ghana, Moses Kofi Sam, for organizing logistics and field visits, and for being of great assistance in every possible way. Many thanks to management and field staff in each of the sites, who


Chomba et al.

Notes on black rhino mortalities in North Luangwa National Park, Zambia

Chansa Chomba,* David Squarre and Harvey Kamboyi

Zambia Wildlife Authority, Directorate of Research, Planning, Information and Veterinary Services, PB 1 Chilanga, Zambia* corresponding author email: [email protected]; [email protected]

In May 2010 reintroduction of the black rhino founder population in North Luangwa National Park (NP) was completed with the release of 5 animals, bringing the total to 25. The rhinos were released in a fenced rhino sanctuary measuring 220 km2 compartmentalized into 55 km2 and 165 km2. Between 2009 and July 2010 4 rhinos died from injuries from intraspecific fights, failure to adapt and trypanosomiasis, reducing the number to 21. During the same period newly born calves increased the population to 30 rhinos, a rate greater than the yearly rate of increase of 5% recommended by the Southern African Development Community for Rhino Conservation. During the first two weeks of November 2011, 6 rhinos died from injuries caused by intraspecific fights and disease, which reduced the number to 24.

The records at Zambia Wildlife Authority (ZAWA) indicate that there are no definite figures for black rhino population density for North Luangwa NP and surrounding Game Management Areas (Table 1).

It is generally accepted that black rhino density should not exceed 0.1 rhinos/km2. It was for this reason that before black rhinos were reintroduced in Zambia, Kelvin Dunham was engaged to evaluate areas in which to reintroduce them. Dunham (2001) recommended a density of 0.1 rhinos/km2 as being the most reasonable stocking rate for rhino reintroduction in North Luangwa NP. This low figure was chosen to eliminate any ecological danger that rhinos confined in a fenced area might be over-stocked.

Based on the recommendations provided by

Dunham (2001), ZAWA in collaboration with the Frankfurt Zoological Society (FZS) formulated a plan to reintroduce black rhinos in North Luangwa NP. The plan provided for the release of a founder population of 20 animals in a fenced area. It was recommended that the adjoining fences be removed to create a single fenced sanctuary large enough to accommodate 20 founders and their offspring.

In 2011, despite the loss of four rhinos through disease, failure to adapt and injury arising from intra-specific fights, the total number of rhinos increased to 30 in the fenced area. In the interim, ZAWA and FZS prepared a comprehensive rhino management plan for North Luangwa NP which, among other things, recommended a partial removal of the fence to allow the animals to disperse over a wider area and prevent unnecessary mortality that might arise from intraspecific fights. By November 2011, just before the rhino plan was approved and implemented, 6 rhino mortalities were recorded between 4 and 13 November 2011.

On 2 November 2011 the carcass was discovered of a 6-year-old female code-named Buntungwa. Scavengers had consumed much of the body, leaving the head with horns and a few other parts. Eight weeks before Buntungwa died Dr Chap Masterson of the International Rhino Foundation had rated its body condition as very good. The presence of horns on the skull and the absence of any bullet marks on the skin ruled out the possibility of poaching. Its calf, which had still depended on it, also died.

RHINO / FIELD NOTES


Rhino / Field notes

On 4 November 2011 an 18-month-old calf of a female code-named Twatemwa was found dead. The carcass had horns but there were extensive horn injuries on the skin. Deep wounds were recorded in the groin area, rib cage, right armpit, lower jaw and front legs (Fig. 1). It was assumed that the calf was killed by a mature male trying to mate with the mother, Twatemwa.

On 10 November 2011 a 7-year-old female code-named Ithala was found dead. This female originated from Ithala Game Reserve, a tsetse-free (Glossina spp.) area in South Africa. Upon release in North Luangwa NP, which is a tsetse-infested area, the rhino started showing signs of trypanosomiasis, a disease caused by a protozoan of the genus Trypanosoma, which is spread by blood-sucking tsetse flies. A trypanosomiasis prophylactic, Samorin, was administered in June 2010. In April 2011 Ithala showed loss of body condition and supplementary feeding was initiated. On 29 August 2011 the animal was immobilized and fitted with a VHF radio transmitter. At the time of fitting the transmitter, the body condition rating was 3 out of 5, which was fair. Her teeth were examined and one of the molars had just erupted out of the gum; this was assumed to be the cause of the fair instead of good or very good body condition rating. On 7 November 2011 rhino monitoring staff recorded poor body condition with clinical signs of trypanosomiasis: observed drowsiness, drooping head, stumbling when

walking and sometimes falling over. On 10 November 2011 the rhino was found dead and post-mortem was done. Post-mortem results showed lower jaw abscess, broken molar, gum inflammation, hydro-pericardium (fluid around the heart), enlarged adrenal glands, worm infestation and general paleness of the carcass. The cause of death was recorded to be inability to browse due to painful gum, fluids around the heart and heavy worm burden (Fig. 2).

Table 1. Black rhino population estimates in Luangwa Valley, Zambia

Author Year Method Area covered Density/km2 Total Remarks

Ansell, W.F.H. 1969 Field patrol sighting records, other records

Luangwa Valley 12,000 Total population for all NPs and game management areas

Caughley, G. 1973 Aerial North and South Luangwa NPs

0.03 2,800 Total was for South Luangwa NP

Kuper, J.H. 1975 Aerial Luangwa Valley 0.01 for North Luangwa NP

445 Total was for North Luangwa NP

Douglas–Hamilton I. et al.

1979 Aerial 20,000 km2 of Luangwa Valley

0.31

Leader-Williams, N.

1985 Dung piles, tracks, scrapes per km2 observed by scouts

South Luangwa NP

0.4

Leader-Williams, N.

1988 Individual recognition, sightings by scouts

200 km2 of South Luangwa NP

0.39 77 (66 adults and 11 calves)

Figure 1. Twatemwa’s carcass showing injuries sustained from fighting.

©C

hans

a C

hom

ba


Chomba et al.

On 13 November 2011 a 14-year-old bull, which was released in 2006, was found dead. This male had always been involved in fights as was observed from the scars on the skin. It was assumed that the animal died from injuries from intraspecific fights.

During the same period, a calf of less than one year old was not sighted and is assumed to be dead.

Chomba and Matandiko (2012) assert that the reintroduction of rhinos in Zambia has been a success and recommend securing additional financial support and increased manpower when the rhinos are released from the fenced sanctuary to the unfenced area of North Luangwa NP. It is now evident that it is not sufficient to emphasize law enforcement alone but biological management of the species as well. Records from observations made on the reintroduced black rhino population in North Luangwa NP suggest that appropriate biological management, including measures to prevent overstocking, are critical. All the 10 rhino mortalities recorded in the park between January 2009 and November 2011 were associated with biological management and not poaching. The continued confinement of the North Luangwa population in a fenced area may reduce the population’s yearly rate of increase to < 5%. ZAWA management should carefully balance its desire to encourage population growth by spreading rhinos to new areas outside the current fenced sanctuary while at the same time securing the species from poaching incursions in less secure areas.

ReferencesAnsell, W.F.H. 1969. The black rhinoceros in Zambia.

Oryx 10:176–192.Chomba, C. and Matandiko, W. 2012. Report on

population status of black and white rhinos in Zambia. Chilanga, Zambia: Zambia Wildlife Authority.

Caughley, G. 1973. Game management. In: Naylor, J.N., Caughley, G., Abel, N.O. and Liberg, O., eds., Luangwa Valley Conservation and Development Project. Report to UNDP/FAO, Rome. pp. 50–157.

Douglas-Hamilton, I., Hillman A.K.K., Holt, P. and Ansell, P. 1979. Luangwa Valley elephant, rhino, and wildlife survey. Report to IUCN/WWF/NYZS. Unpublished.

Dunham, K.M. 2001. Reintroduction of black rhino in the Luangwa Valley, Zambia. Evaluation of areas for the reintroduction of rhinos in Zambia. SADC Regional Programme for Rhino Conservation.

Kuper, J.H. 1975. Elephant and black rhinoceros census in the Luangwa Valley. Zambia Chinzombo Research Station, Mfuwe. Unpublished.

Leader-Williams, N. 1985. Black rhino in South Luangwa National Park: their distribution and future protection. Oryx 19:27–33.

Leader-Williams, N. 1988. Patterns of depletion in black rhino population in Luangwa Valley, Zambia. African Journal of Ecology 26:181–187.

Figure 2. Post-mortem results of Ithala showing enlarged heart muscle (left) and worm infestation (right).

©C

hans

a C

hom

ba


Rhino / Field notes

New maps representing the historical and recent distribution of the African species of rhinoceros: Diceros bicornis, Ceratotherium simum and Ceratotherium cottoniKees Rookmaaker 1* and Pierre-Olivier Antoine 2

1 Editor, Rhino Resource Center (www.rhinoresourcecenter.com), and Senior Research Fellow, Department of Biological Sciences, National University of Singapore, 14 Science Drive 4, Singapore 117543* corresponding author email: [email protected] Laboratoire de Paléontologie, Institut des Sciences de l’Évolution (ISE-M, UMR-CNRS 5554), c.c. 64, Université Montpellier 2, Place Eugène Bataillon, F-34095 Montpellier Cedex 05, France

AbstractThe paper presents new and updated maps showing the historical and current range of the black, white and Nile rhinoceros. Distribution of the species is based on written and iconographical records, as well as museum specimens. The historical range reflects the situation from ad 1500 onwards (or the last part of the Holocene). The current range is divided into original and introduced populations.

RésuméCet article présente de nouvelles cartes mises à jour qui montrent l’habitat historique et actuel du rhinocéros noir, blanc et du Nil. La distribution de l’espèce est basée sur des rapports écrits et iconographiques ainsi que sur des spécimens dans les musées. L’habitat historique reflète la situation à partir de l’année 1500 et après (c’est-à dire la dernière partie de l’Holocène). L’habitat actuel est divisé d’après des populations originelles et introduites.

Introduction

It is common practice in zoological encyclopaedias or monographs to provide maps of the historical and current distribution of a particular animal. These maps are extremely useful to show how widely a species ranged and to visualize the often extreme reduction of the areas where it can be found. This procedure is a powerful and useful conservation tool, because status figures can only be displayed in graphs that might need some mathematical background to be properly understood.

A survey of the available maps of the historical distribution of any of the six species of rhinoceros still extant reveals discrepancies in the interpretation of the data as well as a lack of definition of what is meant by historical. Some would say it is the range as it existed in 1700, or 1800, or a similar general date, without examining in too much detail why that particular year is chosen to be representative.

Our work with historical publications about the rhinoceros for several decades makes it abundantly obvious that the full potential of the old data has yet to be harvested. The sources are mostly available on the Rhino Resource Center (www.rhinoresourcecenter.com), but it takes time and major effort to extract all information on past distribution. The data to be analysed of course include any mention in a published work or unpublished report that the species was seen or shot. Besides this, there are specimens with localities in museums and private collections, works of art including paintings, drawings and rock engravings, place names referring to the animal, and an array of other sources. If funding would ever be available, a historical atlas of the distribution records of all species of rhinoceros would be an incredible asset.

In this paper we present new maps of the historical distribution of the three extant species of rhinoceros in Africa, i.e. the black rhinoceros Diceros bicornis (Linnaeus, 1758), the white rhinoceros Ceratotherium


Rookmaaker and Antoine

simum (Burchell, 1822) and the Nile rhinoceros Ceratotherium cottoni (Lydekker, 1908). C. cottoni was recently elevated to species rank by Groves et al. (2010), but the maps would of course be the same if the previous subspecific arrangement is preferred. The subspecies of the black rhinoceros (Rookmaaker 2011) are not separated in the map of Diceros bicornis.

Sources of dataToday in conservation circles the most commonly consulted maps are those contained in the action plan by Emslie & Brooks (1999), which for the black rhino is copied from Cumming et al. (1990) and for the two then recognized subspecies of the white rhino is a combination of maps by Cumming et al. (1990) and Hillman-Smith et al. (1986). These historical maps are stated to represent the situation around 1700. Previous continent-wide maps or surveys containing a wealth of individual data are those by Zukowsky (1965) and Sidney (1965).

The oldest literary records available on the rhinoceros refer to South Africa from the middle of the 17th century. These were first comprehensively investigated and placed in an ecological context in the remarkable volumes by Jack Skead (1912–2006), recently edited as Skead et al. (2007, 2011). Rookmaaker (2001, 2002) discussed the near-extinction of the white rhino in the early 20th century. The extinction of the species in Zimbabwe was first highlighted by Rookmaaker (2003). All available information on rhinos seen or shot or studied in the period from 1795 to 1875 were studied by Rookmaaker (2008) and the results plotted on maps covering Botswana, Namibia, South Africa, Swaziland and Zimbabwe.

Yalden et al. (1986) and Rookmaaker & Kraft (2011) reviewed early records from Ethiopia and Somalia. Kingdon (1979) produced a remarkable historical map of distribution of rhinos in Kenya. Schomber (1966) and Hillman-Smith et al. (1986) studied the situation in Central Africa as regards the Nile rhinoceros.

The relatively restricted range in countries of West Africa was reviewed by Rookmaaker (2004). A specimen recently discovered by Antoine & Rookmaaker (2012) has extended the distribution of the black rhino into northern parts of Togo.

ResultsIn our view, there is only one realistic way to portray the historical distribution of a particular species. Because records of even a large and prominent mammal like a rhinoceros are often relatively few in number, we must combine everything that might have relevance to their range. A map of historical distribution is therefore a representation of all records of the existence of the species, where it is assumed that if it is known to exist in a certain place in, say, 1850, or 1900, or 1950, or 2000, it also used to exist there at any time before such a date. If for instance, a rhinoceros would have been seen in 2010 in the montane forest on the east side of Mount Kenya, it is assumed that it is one of a population that always lived there. The map then draws a line around all these records that have been plotted on a map, and this we call the historical distribution or range of that species.

There is no particular known or defined measure as to how far back the historical range of a certain extant mammal can be known. Most written records of course refer to the period starting from the start of printed records, from around 1500. In this case, therefore, the historical element of our maps shows the range of the various species in the past five centuries, coinciding with the last stages of the Holocene.

There are, of course, plenty of pitfalls that make the construction of a historical map less easy than it seems (Skead 1962; Boshoff & Kerley 2010). Species distributions change in time and place, and animals move around. It is (almost) obvious that a rhinoceros only lives in places where the habitat is favourable, unless it is moving from one place to another. It is (almost) obvious that a rhinoceros would not be found close to human habitations unless it was left undisturbed. Rhinos could potentially be found on one side of a mountain and not on another, or on one side of a river and not on the other. Such relatively minor and very local distribution patterns can never be shown on continental maps of a historical distribution area and are therefore always ignored.

A map of historical distribution is generally assumed to reconstruct what may be called the original range of a species. Hence any area into which a species of rhino was actively introduced—usually in the period after 1960—that by chance lay outside the known and verified range would not be shown in this kind of map. It is therefore extremely important


Rhino / Field notes

be present, and their placement emphatically has no relationship to the actual range or status. O is used in case of countries, where as far as we know the species has always existed within the past five centuries (or longer); + is used in case of countries—in a few cases parts of countries—where at one time the species was extinct, hence where all current populations have been reintroduced, or (where applicable) introduced.

The three maps showing the historical distribution of the black rhino, the white rhino and the Nile rhino are presented here (Figs. 1–3). They were first published by Antoine & Rookmaaker (2011) in a slightly different format, without, however, providing background into their preparation. It is envisaged to make these maps available on the Rhino Resource

to keep track of all translocations, reintroductions and introductions, although it is likely that the existing literature would not be adequate for this purpose. There are clear guidelines regarding the movement of rhino in continental Africa (Emslie et al. 2009) and it would be advisable to ensure that a full knowledge of historical records is part of all management policies.

Indications of the current distribution are based on the figures and notes in Emslie & Brooks (1999) and updated in Emslie (2008). The current data have been separated into two sets, which are indicated on the maps by two distinct symbols: ‘O’ for original and ‘+’ for (re)introduced. For security reasons, these symbols are placed randomly within the boundaries of the countries in which the species is believed to

Figure 1. Historical and current distribution of the black rhinoceros Diceros bicornis (Linnaeus, 1758). The historical range in the period after ad 1500 is shown in grey. The current range is shown by the symbols ‘O’ for remnants of original populations and ‘+’ for introduced or reintroduced populations.



Center website (www.rhinoresourcecenter.com) where users can view and download them to be used elsewhere with proper acknowledgements.

Acknowledgements

This paper is part of the ongoing research of the Rhino Resource Center, sponsored by SOS Rhino, International Rhino Foundation, WWF Areas, Rhino Carhire and Save the Rhino International.

ReferencesAntoine, P.O. and Rookmaaker, L.C. 2012. The first

historical record of a rhinoceros in Togo. Mammalia, Paris 2012. DOI: 10.1515/mammalia-2012-0049.

Boshoff, A.F. and Kerley, G.I.H. 2010. Historical mammal distribution data: How reliable are written records? South African Journal of Science 106(1/2):1–8.

Cumming, D., Toit, R. du and Stuart, S.N. 1990. African elephants and rhinos: Status survey and conservation action plan. Gland: IUCN. pp. i–iv, 1–72.

Emslie, R.H. 2008. Rhino population sizes and trends. Pachyderm 44:88–95, tables.1–4.

Figure 2. Historical and current distribution of the white rhinoceros Ceratotherium simum (Burchell, 1822). The historical range is shown in grey. The only original population remained in the early 20th century in parts of Zululand (KwaZulu-Natal, South Africa), and all other specimens now found on the African continent in wild or semi-wild conditions have been restocked from the Zululand remnants.


Rhino / Field notes

Emslie, R.H., Amin, R. and Kock, R. 2009. Guidelines for the in situ reintroduction and translocation of African and Asian rhinoceros. Occasional Paper of the IUCN Species Survival Commission no. 39: i–v, 1–115.

Emslie, R.H. and Brooks, M. 1999. African rhino: Status survey and conservation action plan. Gland and Cambridge: IUCN/SSC African Rhino Specialist Group. pp. i–x, 1–92.

Groves, C.P., Fernando, P. and Robovsky, J. 2010. The sixth rhino: A taxonomic re-assessment of the critically endangered northern white rhinoceros. PLoS One 5(4) e9703:1–15.

Hillman-Smith, K., Mankoto ma Oyisenzoo and Smith, F. 1986. A last chance to save the northern white rhino? Oryx 20(1):20–26, figs. 1–3, map 1, table 1.

Kingdon, J. 1979. East African mammals, vol. 3, part B: Large mammals. London: Academic Press. pp. i–iv, 1–436.

Rookmaaker, L.C. 2001. The alleged population reduction of the southern white rhinoceros (Ceratotherium simum simum) and the successful recovery. Säugetierkundliche Mitteilungen 45(2):55–70, fig.1, tables. 1–2.

Rookmaaker, L.C. 2002. Miscounted population of the southern white rhinoceros (Ceratotherium simum simum) in the early 20th century? Pachyderm 32:22–28, fig. 1, tables 1–2.

Rookmaaker, L.C. 2003. The last white rhinoceroses in Zimbabwe. Pachyderm 35:100–114, figs. 1–8.

Figure 3. Historical and current distribution of the Nile rhinoceros Ceratotherium cottoni (Lydekker, 1908). The historical range is shown in grey. It is believed that by 2010 no animals remained in the wild (last seen in NE Congo).



Rookmaaker, L.C. 2004. Historical distribution of the black rhinoceros (Diceros bicornis) in West Africa. African Zoology 39(1):63–70, figs. 1–6.

Rookmaaker, L.C. 2008. Encounters with the African rhinoceros: A chronological survey of bibliographical and iconographical sources on rhinoceroses in southern Africa from 1795 to 1875. Reconstructing views on classification and changes in distribution. Munster, Schuling Verlag. 1–148, figs. 1–157 [68 in colour], maps A–I [2 in colour], tables 1–47 [also: Transactions of the Royal Society of South Africa 62(2):55–198].

Rookmaaker, L.C. 2011. A review of black rhino systematics proposed in ungulate taxonomy by Groves and Grubb (2011) and its implications for rhino conservation. Pachyderm 50:72–76.

Rookmaaker, L.C. and Kraft, R. 2011. The history of the unique type of Rhinoceros cucullatus, with remarks on observations in Ethiopia by James Bruce and William Cornwallis Harris (Mammalia, Rhinocerotidae). Spixiana, Munchen 34(1):133–134, figs. 1–8.

Schomber, H.W. 1966. Die Verbreitung und der Bestand des zentralafrikanischen Breitmaul-nashorns, Ceratotherium simum cottoni (Lydekker, 1908).

Säugetierkundliche Mitteilungen 14:214–227, figs. 1–4, tables 1–12.

Sidney, J. 1965. The past and present distribution of some African ungulates. Transactions of the Zoological Society of London 30:1–397, figs. 1–39, maps 1–9.

Skead, C.J. 1962. The use of early historical references in the study of bird and mammal distribution. Annals of the Cape Provincial Museums 2:138–142.

Skead, C.J., Boshoff, A.F., Kerley, G.I.H. and Lloyd, P.H. 2007. Historical incidence of the larger land mammals in the broader Eastern Cape, 2nd ed. Port Elizabeth: Centre for African Conservation Ecology, Nelson Mandela Metropolitan University. i, i–xiii, 1–570 pp.

Skead, C.J., Boshoff, A.F., Kerley, G.I.H. and Lloyd, P.H. 2011. Historical incidence of the larger land mammals in the broader northern and western Cape, 2nd ed. Port Elizabeth: Centre for African Conservation Ecology, Nelson Mandela Metropolitan University. i, i–xiv, 1–519 pp.

Yalden, D.W., Largen, M. and Kock, D. 1986. Catalogue of the mammals of Ethiopia, part 6. Monitore Zoologico Italiano 21(4):31–103.

Zukowsky, L. 1965. Die Systematik der Gattung Diceros Gray, 1821. Zoologische Garten 30:1–178, figs. 1–8.


Density-dependent effect affecting elephant seed-dispersed tree recruitment

Density-dependent effect affecting elephant seed-dispersed tree recruitment (Irvingia gabonensis) in Congo Forest

David Beaune,1,2,* Loïc Bollache, 2 Barbara Fruth,1 Gottfried Hohmann1 and François Bretagnolle 2

1 Max Planck Institute for Evolutionary Anthropology, Department of Primatology, Deutscher Platz 6, Germany2 Laboratoire Biogéosciences, UMR CNRS 6282, Université de Bourgogne, 6 bd Gabriel, 21000 Dijon, France* corresponding author email: [email protected]

Introduction

Several tree species are known to be important for local wildlife, rural communities (White & Abernethy 1997) and even the ‘Western world’. However, little is known about the ecology of these species, and a biodiversity crisis could change the population survival. Among these, the bush mango (Irvingia gabonensis), widespread in West and Central Africa, is of major importance for rural communities (Atangana et al. 2001; Leakey et al. 2005). Recently, the plant is used as a slimming supplement in the Western world. Elephants are widely recognized as the most important Irvingia seed dispersers in Africa (Theuerkauf et al. 2000; Nchanji & Plumptre 2003; Morgan & Lee 2007). In this study we focus on I. gabonensis as the example to illustrate seed fate without dispersion, and thus density-dependence effect affecting tree recruitment.

We investigated this megafaunal tree population’s ability to survive without elephants in the evergreen lowland rainforest of the Max-Planck research site, LuiKotale, on the south-western fringe of Salonga National Park (NP), Democratic Republic of the Congo (DRC). In and around Salonga NP elephants (Loxondota cyclotis) have been severely poached for decades (Van Krunkelsven et al. 2000; Blake et al. 2007;), and poaching has increased with increasing post-war availability of automatic weapons (AK47) and ammunition. The current nationwide elephant population is said to have declined by as much as two-thirds to that of the 1990s, and the remainder is said to survive in fragmented subpopulations (Alers et al. 1992). Throughout 10 years of continuous

presence at the research site at LuiKotale, pressure on the species became evident when carcasses from massacres were documented.

Overall, we aimed to assess the ability of the I. gabonensis tree community at LuiKotale to reproduce without elephant dispersal. If megafaunal trees depend on elephants for seed dispersal, one would expect no alternative seed dispersers and thus a high mortality of seedlings and poles due to the density-dependent effect (Paine et al. 2012).

Methods

Study siteThe LuiKotale research site is located within the equatorial rainforest 2°47′ S–20°21′ E, at the south-western fringe of Salonga NP (DRC), in the

Figure 1. Sampling area, red spots representing adult trees.


Beaune et al.

same continuous forest block (Fig. 1). This park, classified as a World Heritage Site, is the largest protected rainforest area in Africa and the second largest protected rainforest in the world (33,346 km²) (Grossmann et al. 2008). The study site is a primary evergreen tropical lowland rainforest ancestrally owned and used by Lompole village, 17 km away. The site covers > 60 km² with a network trail of 76 km. Since 2001, all exploitation within the site ceased, to allow research (Hohmann & Fruth 2003).The climate is equatorial with abundant rainfall (> 2000 mm/yr), a small dry season in February and a larger one between May and August. Mean temperature at LuiKotale ranges between 21 °C and 28 °C, with a minimum of 17 °C and a maximum of 38 °C (2007–2010). Five major vegetation types were distinguished: 1) mixed tropical forest on terra firme, 2) monodominant forest dominated by Monopetalanthus sp., 3) monodominant primary forest dominated by Gilbertiodendron dewevrei, 4) temporarily inundated mixed forest, and 5) permanently inundated mixed forest. Dry habitats (1–3) dominate site cover, with 73% heterogeneous and 6% homogenous in composition. Wet habitats (4 + 5) represent 17% and 4% of the cover respectively (Mohneke & Fruth 2008).

To investigate the density-dependent effect on seed survival of this elephant-dependent tree, we focused on all adult trees of Irvingia gabonensis inventoried

since 2007 that produced ripe fruits during the survey from January 2010 to June 2011. (A database of the LuiKotale research site geo-references all feeding trees within the observed range of bonobo, Pan paniscus, communities.)

We counted 1) seeds 2) seedlings 3) saplings and 4) poles in the fruit-fall zone of each individual and judged the state of each of the four stages of growth, assessing pathogens and folivores by visual inspection (absence/presence of traces).

ResultsWe investigated 54 adult trees of I. gabonensis (83.1 cm ± SE. 0.7 diameter at breast height) producing ripe fruit. Figure 2 shows the presence and state of 1) seeds, 2) seedlings, 3) saplings, and 4) poles.

SeedsSeeds were present within all fruit-fall zones. Seeds revealed a loss rate of 54% ± SE 3 due to seed predation, and among the unopened seeds 76% ± 4 were rotten or showed signs of pathogen attacks. Red river hogs (Potamochoerus porcus) in herds of two to six animals were found responsible for predating on large quantities of seeds, cracking the endocarps (Beaune et al. 2012).

Fruit production(500-5000)

Seed predators:cracked seeds ~54%

density-dependent effects

Pathogens:rotten seeds ~76%

Pathogens:infection traces – 76%

Folivores:folivory traces – 100%

Folivores

Pathogens

Adult recruiting pole – 0%

Adult recruiting sapling – 2%

Adult recruiting seedling – 11%

Adult producing fruits – 100%

Mean recruit= 0

Mean recruit– 0.02

Mean recruit– 1.67

Figure 2. The density-dependent effect of Irvingia gabonensis. There was no recruitment under the parental trees (n = 54).


Density-dependent effect affecting elephant seed-dispersed tree recruitment

SeedlingsOnly 6 (11%) of the 54 trees showed seedling recruitment. All 90 of these seedlings were infested by pathogens or showed traces of folivory whereas some of the other surrounding seedling species and the Irvingia of the nursery did not (unpublished data). Although these adult trees reproduced, no established offspring (i.e. those producing fruit) was found beneath the adults’ crowns. A total of 48% (n = 26) of the fruit-fall zones clearly showed tracks of animals leading to the feeding place.

SaplingsA single sapling recruit (< 2 m high) was found below an adult crown.

PolesNo pole was found below an adult crown.

Conclusion

Our results showed a high mortality for Irvingia seeds and recruits on all levels, with a seed loss of 54% to predation and 76% to pathogens and seedling loss of 100% due to predation and pathogens. These results can be explained by the density-dependent effect, also named the Janzen-Connell effect (Janzen 1970; Connell 1971; Burkey 1994), where the mortality of seeds, eggs or other immobile organisms is correlated with their density, which attracts predators

and pathogens. In the absence of an endozoochoric partner such as the elephant, this putting-all-your-eggs-in-one-basket adaptation is likely to turn out as a maladaptation, unless a tree species has alternative dispersal partners or mechanisms.

In the southern area of the Congo River, bonobos, the second biggest frugivores, are unable to replace elephants as seed dispersers, as the seeds are too large for them to swallow. They may contribute in some cases to dispersal outside the fruit-fall zone by short distance ectozoochoric transport, similar to what rodents can disperse (Forget & Wall 2001). For I. gabonensis, bonobos can be

considered as a poor disperser, dispersing over much shorter distances than elephants and omitting passage through their digestive tract.

The survival of I. gabonensis is compromised without a seed-dispersal vector such as forest elephants.

ReferencesAlers, M.P.T., Blom, A., Kiyengo, C.S., Masunda, T. and

Barnes, R.F.W. 1992. Preliminary assessment of the status of the forest elephant in Zaire. African Journal of Ecology 30:279–291.

Atangana, A.R., Tchoundjeu, Z., Fondoun, J.M., Asaah, E., Ndoumbe, M. and Leakey, R.R.B. 2001. Domestication of Irvingia gabonensis: 1. Phenotypic variation in fruits and kernels in two populations from Cameroon. Agroforestry Systems 53:55–64.

Beaune, D., Bollache, L., Fruth, B. and Bretagnolle, F. 2012. Bush pig (Potamochoerus porcus) seed predation of bush mango (Irvingia gabonensis) and other plant species in Democratic Republic of Congo. African Journal of Ecology 50(4):509–512.

Blake, S., Strindberg, S., Boudjan, P., Makombo, C., Bila-Isia, I., Ilambu, O., Grossmann, F., Bene-Bene, L., de Semboli, B., Mbenzo, V., S’Hwa, D., Bayogo, R., Williamson, L., Fay, M., Hart, J. and Maisels, F. 2007. Forest elephant crisis in the Congo Basin. PloS Biology 5:945–953.

Burkey, T.V. 1994. Tropical tree species diversity: a test of the Janzen-Connell model. Oecologia 97:533–540.

Connell, J.H. 1971. On the role of natural enemies in

Figure 3. Seedling and adult tree of Irvingia gabonensis.


Beaune et al.

preventing competitive exclusion in some marine mammals and in rain forest trees. In: Boer, P.J. and Gradwell, G. (eds.), Dynamics of populations, PUDOC. pp. 298–310.

Forget, P.M. and Wall, S.B.V. 2001. Scatter-hoarding rodents and marsupials: convergent evolution on diverging continents. Trends in Ecology and Evolution 16:65–67.

Grossmann, F., Hart, J.A., Vosper, A. and Ilambu, O. 2008. Range occupation and population estimates of bonobos in Salonga National Park: Application to large-scale surveys of bonobos in the Democratic Republic of Congo. In: Furuichi, T. and Thompson, J. (eds.), Bonobos: behavior, ecology and conservation. New York: Springer Science+Business Media, pp. 189–216.

Hohmann, G. and Fruth, B. 2003. LuiKotal—A new site for field research on bonobos in the Salonga National Park. Pan Africa News 10:25–27.

Janzen, D.H. 1970. Herbivores and the number of tree species in tropical forests. American Naturalist 104:501.

Leakey, R.R.B., Greenwell, P., Hall, M.N., Atangana, A.R., Usoro, C., Anegbeh, P.O., Fondoun, J.M. and Tchoundjeu, Z. 2005. Domestication of Irvingia gabonensis: 4. Tree-to-tree variation in food-thickening properties and in fat and protein contents of dika nut. Food Chemistry 90:365–378.

Mohneke, M. and Fruth, B. 2008. Bonobo (Pan paniscus) density estimation in the SW Salonga

National Park/Democratic Republic of Congo: Common methodology revisited. In: Furuichi, T. and Thompson, J. (eds.), Bonobos: Behavior, ecology, and conservation. pp. 151–166.

Morgan, B.J. and Lee, P.C. 2007. Forest elephant group composition, frugivory and coastal use in the Reserve de Faune du Petit Loango, Gabon. African Journal of Ecology 45:519–526.

Nchanji, A.C. and Plumptre, A.J. 2003. Seed germination and early seedling establishment of some elephant-dispersed species in Banyang-Mbo Wildlife Sanctuary, south-western Cameroon. Journal of Tropical Ecology 19:229–237.

Paine, C.E.T., Norden, N., Chave, J., Forget, P-M., Fortunel, C., Dexter, K.G. and Baraloto, C. 2012. Phylogenetic density dependence and environmental filtering predict seedling mortality in a tropical forest. Ecology Letters 15:34–41.

Theuerkauf, J., Waitkuwait, W.E., Guiro, Y., Ellenberg, H. and Porembski, S. 2000. Diet of forest elephants and their role in seed dispersal in the Bossematie Forest Reserve, Ivory Coast. Mammalia 64:447–459.

Van Krunkelsven, E., Bila-Isia, I. and Draulans, D. 2000. A survey of bonobos and other large mammals in the Salonga National Park, Democratic Republic of Congo. Oryx 34:180–187.

White, L. and Abernethy, K. 1997. A guide to the vegetation of the Lope Reserve, Gabon. New York: Wildlife Conservation Society.


CITES-MIKE update

MIKE 3.0As reported in Pachyderm 51, the second phase of the MIKE programme in Africa, which benefitted from generous funding from the European Commission since 2007, concluded at the end of 2012. There was considerable uncertainty as to whether funding could be secured to continue MIKE operations beyond that date, but on 7 November 2012 the European Commission approved a new grant under its Thematic Programme for the Environment and Natural Resources (ENRTP), which will allow MIKE to continue for a further two years. The CITES Secretariat is extremely grateful to the European Union for its continued support to the MIKE programme.

This new phase, dubbed MIKE 3.0, aims to consolidate the achievements of the previous two phases, taking advantage of key lessons learned to ensure that the ranger-based monitoring systems implemented at sites are consistent with site-level capacity. MIKE 3.0 also seeks to establish sustainable mechanisms for training in collaboration with appropriate training institutions. In pursuit of sustainability, MIKE 3.0 will work to ensure that monitoring systems are relevant to conservation and adaptive management requirements at all levels, from field sites through national authorities to subregional institutions and the international community at large.

As with MIKE phases 1 and 2, IUCN

CITES-MIKE updateMise à jour de la CITES-MIKE

Julian Blanc

Acting Coordinator and Data Analyst, CITES-MIKE, UNEP/DELC, United Nations Compound T-36, UN Avenue Gigiri, Nairobiemail: [email protected]

MIKE 3.0Comme indiqué dans Pachyderm 51, la deuxième phase du programme MIKE en Afrique qui a bénéficié d’un financement généreux de la Commission Européenne depuis 2007, a pris fin à la fin de 2012. Il y avait beaucoup d’incertitude quant à savoir si le financement pourrait être assuré pour poursuivre les activités de MIKE au-delà de cette date, mais le 7 novembre 2012, la Commission Européenne a approuvé une nouvelle subvention sous son Programme thématique pour l’Environnement et la gestion des Ressources Naturelles (ENRTP), qui permettra à MIKE de continuer pendant encore deux ans. Le Secrétariat de la CITES est extrêmement reconnaissant à l’Union Européenne pour son soutien continu au programme MIKE. Cette nouvelle phase, baptisée MIKE 3.0, vise à consolider les acquis des deux premières phases, en profitant des principales leçons apprises afin de s’assurer que les systèmes de surveillance par les écogardes, mis en œuvre sur les sites sont compatibles avec les capacités au niveau du site. MIKE 3.0 cherche également à établir des mécanismes durables de formation en collaboration avec des institutions de formation appropriées. Dans la poursuite de la durabilité, MIKE 3.0 va veiller à ce que les systèmes de contrôle soient pertinents à la conservation et aux exigences de la gestion adaptative à tous les niveaux, à partir des sites sur le terrain en passant par les autorités nationales aux institutions sous-régionales et la communauté internationale dans son ensemble. Comme pour les phases 1 et 2 de MIKE, l’UICN sera le principal partenaire de MIKE au niveau de la mise en œuvre sur le terrain en hébergeant les unités d’appui

MIKE–ETIS UPDATES


Blanc

will be MIKE’s principal partner at the field implementation level through hosting the MIKE subregional support units. However, and in line with the current economic climate, funding for MIKE 3.0 is at a considerably reduced level compared to the previous phases of the programme. This has inevitably resulted in some changes, including the loss of some loyal and valuable members of the MIKE team. Sani Massalatchi—MIKE subregional support officer (SSO) for West Africa—left the programme on 30 October 2012, while Edison Nuwamanya—SSO for Eastern Africa—left on 31 December. Saying goodbye is never easy, and particularly to colleagues who have invested a substantial part of their professional lives to the MIKE programme. The CITES Secretariat and its MIKE programme are grateful to Edison and Sani for their invaluable contributions over the years and wish them the very best in all their future endeavours.

As a result of these changes, the previous four sub-regional support units (SSUs) are being consolidated into two units. The Eastern and Southern Africa SSUs are now merged into a single office, based in Pretoria, South Africa, and overseen by Tapera Chimuti with administrative support by Ditse Mduli. The Central and West Africa office will be headed by Sébastien Luhunu from Yaoundé, Cameroon, supported by Martha Bechem and with administrative assistance by Solange Soh and Edith Lompo. The MIKE Central Coordinating Unit (CCU), which continues to be hosted by the United Nations Environment Programme in Nairobi, has also been downsized, with the two professional positions that existed in phase 2 (MIKE coordinator and data analyst) now being merged into a single post.

Another feature of MIKE 3.0 will be the progressive extension of monitoring activities to other CITES-listed mammal species threatened by trade. The deployment of the MIST software by MIKE over the last few years, and the forthcoming rollout of SMART (more on which below), has enabled sites to use MIKE-promoted methods to collect data on illegal activities affecting other species of wildlife. As elephants are far from the only species currently threatened by illegal killing and trade, MIKE 3.0 will pave the way for the development of a larger, multi-species project, which will aim not only to monitor levels of illegal

sous-régionaux de MIKE. Cependant, et en accord avec le contexte économique actuel, le financement de MIKE 3.0 est à un niveau considérablement réduit par rapport aux phases précédentes du programme. Cela a inévitablement entraîné certains changements, y compris la perte de certains membres loyaux et précieux de l’équipe de MIKE. Sani Massalatchi, responsable de l’appui sous-régional pour l’Afrique occidentale, a quitté le programme le 30 octobre 2012, tandis qu’Edison Nuwamanya, responsable de l’appui sous-régional pour l’Afrique orientale, est parti le 31 décembre. Dire au revoir n’est jamais facile, et surtout aux collègues qui ont investi une partie importante de leur vie professionnelle au programme MIKE. Le Secrétariat de la CITES et son programme MIKE sont reconnaissants à Edison et Sani pour leur contribution inestimable au fil des années et nous leur souhaitons nos meilleurs vœux dans tous leurs futurs projets.

En raison de ces changements, les quatre unités de soutien sous-régional (USS) précédentes sont regroupées en deux unités. Les USS d’Afrique orientale et australe sont maintenant fusionnées en un seul bureau basé à Pretoria, Afrique du Sud, et supervisé par Tapera Chimuti avec le soutien administratif de Ditse Mduli. Le bureau d’Afrique centrale et occidentale sera dirigé par Sébastien Luhunu de Yaoundé, Cameroun, soutenu par Martha Bechem avec l’assistance administrative de Solange Soh et Edith Lompo. L’UCC MIKE, qui continue à être abritée par le Programme des Nations Unies pour l’Environnement à Nairobi, a également été réduit, et les deux positions professionnelles qui existaient dans la phase II (coordonnateur de MIKE et analyste de données) sont maintenant fusionnées en un seul poste. Une autre caractéristique de MIKE 3.0 sera l’extension progressive des activités de surveillance à d’autres espèces de mammifères inscrites par la CITES et menacées par le commerce. Le déploiement du logiciel MIST par MIKE au cours des dernières années, et le déploiement prochain de SMART (plus de détails ci-dessous), ont permis aux sites d’utiliser des méthodes promues par MIKE pour collecter des données sur les activités illégales affectant d’autres espèces de la faune. Comme les éléphants sont loin d’être les seules espèces actuellement menacées par le braconnage et le commerce, MIKE 3.0 va ouvrir la voie à l’élaboration d’un plus grand projet couvrant plusieurs espèces, qui aura pour but non seulement de surveiller les niveaux d’abattage illégal qui affectent une série d’espèces au niveau mondial, mais aussi de renforcer les capacités pour une action de conservation efficace.


CITES-MIKE update

killing in a suite of species at a global level, but also to build capacity for effective conservation action.

Reports to CITES SC62 and CoP16The downsizing of MIKE operations comes at a testing time for elephant populations in Africa. At the 62nd meeting of the CITES Standing Committee in July 2012, the MIKE programme reported that 2011 showed the highest levels of elephant poaching since MIKE records began 10 years earlier. MIKE data suggest that the scale of illegal killing of elephants exceeded sustainable levels in all four African subregions in 2011, meaning that more elephants were killed than were born in that year. If this trend is sustained, it could lead to measurable declines in elephant numbers across the continent. This pattern is matched by the information reported by our sister programme ETIS, with 2011 having the highest volume of ivory seized since the 1989 ivory trade ban.

The reports prepared by MIKE and ETIS for SC62 were integrated in a joint report that also included information on the status of elephant populations provided by the African and Asian Elephant Specialist Groups; details on enforcement actions undertaken to curb the illegal trade in ivory provided by the CITES Secretariat; information on the status of the African Elephant Fund, provided by South Africa on behalf of the African elephant range States; and data on the legal trade in ivory supplied by UNEP-WCMC. The joint report, which is available from http://cites.org/eng/com/SC/62/E62-46-01.pdf, was very well received by the Standing Committee. In response to the report, the Committee adopted a suite of decisions, including a request to the eight countries identified as being most heavily involved in the illegal ivory trade chain to submit written reports to the Secretariat by the end of 2012 on actions taken to stem the flow of illegal ivory through their territories.

These results of the joint report were also presented at a meeting convened and chaired by the CITES Secretary General, Mr John Scanlon, in Brussels, Belgium, in November 2012. The meeting included presentations by Holly Dublin, Chair of AfESG and Steve Broad, Executive

Rapports au CP62 de la CITES et à la CdP 16 La réduction des activités de MIKE intervient à un moment éprouvant pour les populations d’éléphants en Afrique. Lors de la 62ème réunion du Comité permanent de la CITES, le programme MIKE a indiqué que 2011 a vu les niveaux les plus élevés de braconnage d’éléphants depuis que les données de MIKE ont commencé il y a 10 ans. Les données de MIKE suggèrent que l’ampleur de l’abattage illégal des éléphants a dépassé les niveaux viables dans les quatre sous-régions d’Afrique en 2011, ce qui signifie qu’on a tué plus d’éléphants que ceux qui sont nés au cours de cette année-là. Si cette tendance se maintient, elle pourrait conduire à des baisses notables dans le nombre d’éléphants à travers le continent. Cette tendance va de pair avec les informations rapportées par notre programme sœur ETIS, selon lesquelles 2011 a vu le plus grand volume d’ivoire saisi depuis l’interdiction du commerce de l’ivoire en 1989.

Les rapports établis par MIKE et ETIS pour le CP62 ont été intégrés dans un rapport conjoint qui comprenait également des informations sur l’état des populations d’éléphants fournis par les groupes de spécialistes des éléphants d’Afrique et d’Asie, les détails sur les mesures prises pour lutter contre le commerce illégal de l’ivoire fournis par le Secrétariat de la CITES, les informations sur la situation du Fonds pour l’Eléphant d’Afrique fournies par l’Afrique du Sud au nom des Etats de l’aire de distribution des éléphants d’Afrique, et les données sur le commerce légal de l’ivoire fournies par le PNUE-WCMC. Le rapport conjoint, qui est disponible sur le site http://cites.org/eng/com/SC/62/E62-46-01.pdf, a été très bien accueilli par le Comité permanent. En réponse à ce rapport, le Comité a adopté un ensemble de décisions, y compris une demande aux huit pays identifiés comme étant les plus fortement impliqués dans la chaîne du commerce illégal de l’ivoire de soumettre des rapports écrits au Secrétariat avant la fin de 2012 sur les mesures prises pour endiguer le flux d’ivoire illégal sur leur territoire.

Les résultats du rapport conjoint ont également été présentés lors d’une réunion convoquée et présidée par le Secrétaire général de la CITES, M. John Scanlon, à Bruxelles, Belgique, en novembre 2012. La réunion comportait des présentations par Holly Dublin, Présidente du GSEAf et Steve Broad, Directeur exécutif de TRAFFIC International, ainsi que par le programme MIKE. Ont participé à la réunion une délégation des


Blanc

Director of TRAFFIC International, as well as from the MIKE programme. The meeting was attended by a delegation of African elephant range State wildlife authorities (Botswana, Cameroon and Kenya), the Secretariat of the African, Caribbean and Pacific Group of States (ACP), the European Commission, the European Parliament, and representatives of the governments of several European Union member States. In a productive discussion, participants highlighted the important role that the monitoring and joint reporting that MIKE, ETIS and the specialist groups plays in gaining a better understanding of the dynamics of the illegal ivory trade and in elephant conservation in general.

The report submitted by the MIKE programme for consideration at the 16th meeting of the Conference of Parties (Bangkok, Thailand, 3–14 March 2013) is largely based on the report submitted to SC62. A large amount of carcass data from Africa and Asia has been received for the first half of 2012 and is still being collated. These data will be used to produce an updated analysis of trends in the illegal killing of elephants and will be incorporated into the MIKE report to CoP 16. The CoP16 report, available from http://www.cites.org/eng/cop/16/doc/E-CoP16-53-01.pdf, also highlights achievements made in the course of MIKE implementation in the course of its second phase.

SMART: A new-generation site conservation toolMuch of the progress made in MIKE implementation in recent years is a direct consequence of the deployment by MIKE of flexible and user-friendly ranger-based monitoring systems designed to meet first and foremost the monitoring needs of conservation areas. Such systems, which are also able to capture and manage the data needed by MIKE, are not seen by site managers as an external imposition or a reporting burden, but as a genuinely useful tool to aid in site management. This has resulted in encouraging levels of uptake, with no fewer than 16 range States deciding in recent years to adopt such systems beyond their MIKE sites and across their protected area networks.

As reported in Pachyderm 49, the MIKE

autorités de la faune des états de l’aire de répartition de l’éléphant d’Afrique (le Botswana, le Cameroun et le Kenya), le Secrétariat du Groupe des Etats d’Afrique, des Caraïbes et du Pacifique (ACP), la Commission Européenne, le Parlement européen et les représentants des gouvernements de plusieurs Etats membres de l’Union européenne. Au cours d’une discussion productive, les participants ont souligné le rôle important que la surveillance et les rapports conjoints que MIKE, ETIS et les groupes de spécialistes jouent pour promouvoir une meilleure compréhension de la dynamique du commerce illégal de l’ivoire et la conservation des éléphants en général.

Le rapport présenté par le programme MIKE pour examen à la 16ème réunion de la Conférence des Parties (Bangkok, Thaïlande, du 3 au 14 mars 2013) repose en grande partie sur le rapport présenté au Cp62. On a reçu une grande quantité de données sur les carcasses en provenance d’Afrique et d’Asie pour le premier semestre de 2012 et on continue toujours à les rassembler. Ces données seront utilisées pour produire une analyse actualisée des tendances de l’abattage illégal des éléphants et seront incorporées dans le rapport de MIKE présenté à la CdP 16. Le rapport CdP16, disponible sur le site http://www.cites.org/eng/cop/16/doc/E-CoP16-53-01.pdf, met également en lumière les réalisations effectuées dans le cadre de la mise en œuvre de MIKE au cours de sa deuxième phase.

SMART: Un outil de la nouvelle génération pour la conservation sur siteUne grande partie des progrès réalisés dans la mise en œuvre de MIKE ces dernières années est une conséquence directe de la mise en place par MIKE des systèmes de surveillance flexibles et faciles à utiliser basés sur les écogardes conçus pour répondre avant tout aux besoins de la surveillance des aires de conservation. Ces systèmes, qui sont également en mesure de saisir et de gérer les données requises par MIKE, ne sont pas considérés par les gestionnaires des sites comme imposés de l’extérieur ou un fardeau de rapport, mais comme un outil vraiment utile pour aider à la gestion du site. Il en a résulté des niveaux d’adoption très encourageants, avec pas moins de 16 Etats de l’aire de distribution qui ont décidé d’adopter ces dernières années de tels systèmes au-delà de leurs sites MIKE et à travers leurs réseaux d’aires protégées.

Comme indiqué dans Pachyderm 49, depuis quelques années le programme MIKE a été étroitement impliqué


CITES-MIKE update

programme has for the last few years been closely involved in the SMART partnership, a consortium of conservation agencies that aims to develop a new and improved tool for measuring, evaluating and improving the effectiveness of wildlife law-enforcement patrols and site-based conservation activities. At the core of SMART is a cross-platform, free and open source software application to collect, manage and analyse spatial data relevant to conservation area management and protection. Building on the strengths of existing systems, SMART is extensible and plug-ins can be developed for specific applications, such as ecological monitoring, patrol planning and intelligence, human–wildlife conflict monitoring and mitigation, and virtually any other site-level conservation activity that involves events in time and space.

SMART aims to be much more than a data management tool—it is a suite of best practices developed by the people who use them and designed to help protected area and wildlife managers better plan, evaluate and implement their activities and to promote good governance. SMART includes a desktop application, training and implementation manuals, web-based training materials, standardized protocols and an active and growing community of users and conservation practitioners who can share experiences and have a say in improving and sustaining SMART over the long term.

A beta version of the SMART software was released in August 2012 and the full public launch of SMART v.1.0 is scheduled for early 2013. For more details and to download the software and training materials, visit the SMART website at http://www.smartconservationsoftware.org.

dans le partenariat SMART, un consortium d’organismes de conservation qui vise à développer un nouvel outil amélioré pour mesurer, évaluer et améliorer l’efficacité des patrouilles de mise en application de la loi sur la faune et les activités de conservation basées sur le site. Au cœur de SMART, il y a une application logicielle de plusieurs plateformes, gratuit et aux sources ouvertes pour recueillir, gérer et analyser les données spatiales relatives à la gestion des aires de conservation et de protection. En s’appuyant sur les points forts des systèmes existants, SMART est extensible et on peut développer des modules complémentaires pour des applications spécifiques, telles que le suivi écologique, la planification des patrouilles et la collecte des renseignements, le suivi du conflit homme-faune et son atténuation, et pratiquement toute autre activité de conservation au niveau du site qui implique des événements dans le temps et dans l’espace.

SMART vise à être beaucoup plus qu’un outil de gestion des données. Il s’agit d’un ensemble de meilleures pratiques développées par les personnes qui les utilisent, conçues pour aider les gestionnaires des aires protégées et de la faune à mieux planifier, évaluer et mettre en œuvre leurs activités et promouvoir la bonne gouvernance. SMART comprend une application assistée par ordinateur, des manuels de formation et de mise en œuvre, un matériel de formation en ligne, des protocoles standardisés et une communauté active et croissante des utilisateurs et des professionnels de la conservation qui peuvent partager leurs expériences et qui ont leur mot à dire dans l’amélioration et le maintien à long terme de SMART.

Une version bêta du logiciel SMART a été publiée en août 2012 et le lancement public complet de SMART v.1.0 est prévu pour début 2013. Pour plus de détails et pour télécharger le logiciel et le matériel de formation, visitez le site Web de SMART sur http://www.smartconservationsoftware.org.


Milliken and Sangalakula

The last six months have been a busy time for ETIS. With the 16th meeting of the CITES Con-ference of Parties (CoP16) rapidly approaching, the fifth comprehensive analysis of the ETIS data has been completed and is now available on the CITES website (http://www.cites.org/eng/cop/16/doc/E-CoP16-53-02-02.pdf). The ETIS data rep-resent the largest collection of seizure records on elephant product trade in the world. As of 26 June 2012 the data comprised 18,302 elephant product seizure records from 89 countries or territories, compiled since 1989. Since the ETIS analysis to CoP15 in 2009, nearly 4,000 additional elephant product seizure records have been added to the database, but 2012—with only 129 seizure re-cords—remained data deficient. Thus, the CoP16 analysis was restricted to the 16-year period 1996 through 2011.

This analysis breaks new ground and introduces a number of refinements thanks to a Darwin Initiative grant from the UK government with the University of Reading that allowed statisticians Dr Fiona Underwood and Robert W. Burn to develop a new analytical framework for the ETIS data. While the basic conceptual design previously used for all ETIS analyses since CoP12 still underpins this assessment, using Bayesian hierarchical modelling, a set of new or improved features now characterize the refined methodological approach. These include:• superior bias-adjustment methods to account

for varying seizure and reporting rates by individual countries in each year

• better smoothing techniques to reduce anomalies that are not indicative of underlying patterns of

Progress in implementing the Elephant Trade Information System (ETIS)Avancement dans la mise en œuvre du Système d’Information sur le Trafic des Eléphants (ETIS)

Tom Milliken1 and Louisa Sangalakula2

1 Elephant & Rhino Programme Coordinator, TRAFFIC, PO Box CY 1409, Causeway, Harare, Zimbawe email: [email protected] 2 TRAFFIC East/Southern Africa, PO Box CY 1409, Causeway, Harare, Zimbabwe

Pendant les six derniers mois, ETIS était très occupé. Avec l’approche rapide de la 16ème réunion de la Conférence des Parties de la CITES (CdP16), la cinquième analyse complète des données d’ETIS a été achevée et elle est maintenant disponible sur le site Web de la CITES (http://www.cites.org/eng/cop/16/doc/E-CoP16-53-02-02.pdf). Les données d’ETIS représentent la plus grande collection de données de saisie sur le commerce des produits issus des éléphants dans le monde, et au 26 juin 2012 elles comprenaient 18 302 dossiers de saisies de produits issus des éléphants provenant de 89 pays ou territoires depuis 1989. Depuis l’analyse ETIS à la CdP15 en 2009, près de 4000 dossiers supplémentaires sur la saisie des produits issus des éléphants ont été ajoutés à la base de données, mais l’année 2012 qui a enregistré seulement 129 dos-siers de saisie est restée une année ayant des données insuffisantes. Ainsi, l’analyse de la CdP16 a été limitée à la période de 16 ans, de 1996 à 2011.

Cette analyse fait des innovations et introduit un certain nombre d’améliorations grâce à une subvention de l’Initiative Darwin du Gouvernement britannique à l’Université de Reading qui a permis aux statisticiens Dr. Fiona Underwood et Robert W Burn d’élaborer un nouveau cadre analytique pour les données d’ETIS. Bien que la conception de base précédemment utilisée pour toutes les analyses d’ETIS depuis la CdP12 sous-tende encore cette évaluation, en utilisant la modélisation hiérarchique bayésienne, un ensemble de fonctionnalités nouvelles ou améliorées caractérisent désormais l’approche méthodologique peaufinée. Il s’agit notamment:• Des méthodes supérieures d’ajustement de biais pour

rendre compte des différents taux de saisies et des rapports par les différents pays pour chaque année

• De meilleures techniques de lissage pour réduire les anomalies qui ne sont pas indicatives des tendances du commerce sous-jacentes dans les données


ETIS update

trade in the data• refinements in the assessment of the ivory

seizure data so that raw and worked ivory are considered separately in three weight classes (< 10 kg, 10 kg to < 100 kg, and equal to or > 100 kg) in recognition of the fact that ivory trade dynamics vary considerably by ivory type and the scale of the transaction

• the production of a Transactions Index to produce relative smoothed trends of illegal ivory trade activity, and

• the production of a Weight Index to establish relative, smoothed trends of the weight of illegal ivory over time.

In terms of results, bias adjusted and smoothed data show a sharp escalation in the illicit trade in ivory. Indeed, from 2007 onwards, illegal trade in ivory has surged in each successive year reaching the highest level in at least 16 years in 2011 (Figure 1). The weight of ivory involved in illegal ivory trade is now roughly three times greater than what was occurring in 1998. When these findings from ETIS are viewed together with the results of the CITES MIKE programme, it is

• Améliorations dans l’analyse des données sur les saisies d’ivoire de sorte que l’ivoire brut et travaillé soient considérés séparément dans trois catégories de poids (<10 kg, de 10 kg à <100 kg et égal ou >100 kg), en reconnaissance du fait que la dynamique du commerce de l’ivoire varie considérablement selon le type d’ivoire et l’ampleur de la transaction

• La production d’un indice de transactions afin de produire des tendances lissées relatives à l’activité du commerce illégal de l’ivoire, et

• La production d’un indice de pondération pour établir les tendances relatives lissées du poids de l’ivoire illégal au fil du temps.

En termes de résultats, les données de biais ajustées et lissées montrent une intensification brusque du commerce illicite de l’ivoire. En effet, à partir de 2007, l’on a vu une montée du commerce illégal de l’ivoire dans chaque année successive qui a atteint en 2011 son plus haut niveau depuis au moins 16 ans (figure 1). Le poids de l’ivoire impliqué dans le commerce illégal est maintenant à peu près trois fois plus élevé qu’en 1998. Lorsque ces conclusions d’ETIS sont considérées conjointement avec les résultats du programme MIKE de la CITES, il est de plus en plus évident que les éléphants sont confrontés à

Figure 1. Trend showing relative weight of ivory in illegal trade by weight class and ivory type, 1996–2011 (ETIS, 26 June 2012). Tendance montrant le poids relatif de l’ivoire dans le commerce illégal par catégorie de poids et type d’ivoire, 1996-2011 (ETIS 26 juin 2012)

1996

0

500

1000

1500

1998 2000 2002 2004 2006 2008 2010

Rel

ativ

e w

eigh

t / p

oid

worked / travaillé 100 kg+worked / travaillé 10–100 kgworked / travaillé < 10 kg raw / brut 100 kg+raw / brut 10–100 kgraw / brut <10 kg



increasingly evident that elephants are facing the most serious conservation crisis since the 1989 trade ban was imposed under the convention. It is hoped that these worrying results will spur the CITES parties to take decisive action to curtail illegal trade in ivory.

The sharp upward trend is being driven by a major increase in large-scale movements of ivory, (which is reflected in Figure 1 in the 100+ kg weight class for raw ivory). The largest of these are ivory seizures that involve 800 kg or more in a single consignment; there were 8 such seizures in 2009, 9 in 2010 and 17 in 2011, totalling nearly 61 tonnes of ivory. The increasing pattern of large movements of ivory is the telltale work of sophisticated international criminal syndicates that link Africa with Asia. Currently, the lack of forensic examination to source ivory, the absence of itemized inventory lists concerning the composition of such seizures, and the general failure to conduct dedicated, long-term investigations along the entire trade chain stand as serious impediments to effective law enforcement. It is disappointing to note that few if any arrests occur with respect to most large-scale ivory seizure cases. Failing to prosecute and convict the criminals behind serious ivory trade crime is a key reason that it continues. For this reason, investigation of large-scale ivory seizures should be recognized by the CITES parties as the single most important ivory trade infraction for focused follow-up attention.

In the period 2009 through 2011, the predominant trade pattern typically involves Kenya, Tanzania and South Africa as the primary exit points for ivory from the African continent, which then moves through Hong Kong SAR, Malaysia, Philippines or Viet Nam as the principal transit countries or territories before reaching China and Thailand, the greatest end-use destinations driving the trade. These nine countries and territories, in varying combinations, comprise the paramount trade chain through which the greatest volume of ivory is flowing at the present time.

Another 10 countries and territories—Cameroon, Congo, Democratic Republic of the Congo, Egypt, Ethiopia, Gabon, Mozambique, Nigeria, Taiwan and Uganda—represent a secondary level of concern as they repeatedly

la crise de conservation la plus grave depuis l’interdiction du commerce en 1989 en vertu de la Convention. Il est à espérer que ces résultats inquiétants stimuleront les Parties à la CITES à prendre des mesures décisives pour réduire le commerce illégal de l’ivoire.

La forte tendance à la hausse est causée par une augmentation importante des grands mouvements de l’ivoire (reflétés dans la figure 1 dans la classe 100 + kg d’ivoire brut). Les plus importants sont les saisies d’ivoire de 800 kg ou plus dans un même envoi: il y avait 8, 9 et 17 de telles saisies en 2009, 2010 et 2011 respectivement, pour un total de près de 61 tonnes d’ivoire. La tendance croissante des grands mouvements d’ivoire révèle l’existence d’organisations criminelles internationales sophistiquées qui relient l’Afrique à l’Asie. Actuellement, le manque d’expertise médico-légale pour déterminer la source de l’ivoire, l’absence de listes d’inventaires détaillés concernant la composition de ces saisies, ainsi que l’absence d’enquêtes spéciales à long terme sur toute la chaîne commerciale demeurent des obstacles sérieux à l’application efficace de la loi. Il est décevant de constater que peu d’arrestations, s’il y en a, se produisent à l’égard de la plupart des grandes saisies d’ivoire. Le fait que les criminels derrière le crime grave du commerce de l’ivoire ne sont pas poursuivis et condamnés est l’une des principales raisons que cela continue. Pour cette raison, l’enquête sur les saisies d’ivoire à grande échelle doit être reconnue par les Parties à la CITES comme la plus importante infraction du commerce de l’ivoire nécessitant une attention de suivi ciblée.

Au cours de la période 2009 à 2011, le circuit prédominant du commerce implique généralement le Kenya, la Tanzanie et l’Afrique du Sud comme les principaux points de sortie de l’ivoire en provenance du continent africain, qui passe ensuite par Hong Kong, la Malaisie, les Philippines et/ou le Viet Nam en tant que principaux pays ou territoires de transit avant d’atteindre la Chine et la Thaïlande, qui sont les plus grandes destinations d’utilisation finale régissant le commerce. Ces neuf pays et territoires, dans des combinaisons variables, constituent la principale chaîne commerciale à travers laquelle le plus grand volume d’ivoire circule à l’heure actuelle.

Dix autres pays et territoires, le Cameroun, le Congo, la République Démocratique du Congo, l’Egypte, l’Ethiopie, le Gabon, le Mozambique, le Nigéria, Taiwan et l’Ouganda, représentent un niveau secondaire de préoccupation, car ils ne cessent de jouer un rôle important de soutien au commerce illicite de l’ivoire. A l’exception de l’Egypte et de Taiwan, qui n’ont pas de populations


ETIS update

play important supporting roles in the illicit ivory trade. Except for Egypt and Taiwan, which have no native elephant populations, all of these players are continual sources of illegal ivory, and some also function as entrepôt or transit countries and on occasion as exit points for ivory leaving the African continent. Taiwan is a potential transit link for ivory moving through Asia, while Egypt harbours a major unregulated domestic ivory market. Overall, the 9 priority and 10 secondary countries and territories of concern collectively account for two-thirds of all illicit ivory trade activity since CoP15, according to the ETIS analysis. The presence of organized crime syndicates is an increasingly entrenched feature behind this illicit trade, and governance shortcomings seriously undermine effective law enforcement along the entire trade chain.

The ETIS analysis does not establish a correlation between the first conditional one-off ivory sale under CITES in June 1999 and subsequent illicit trade in ivory. Following that event, the bias-adjusted and smoothed trend shows some incremental increase but remains generally flat overall. On the other hand, the second one-off ivory sale under CITES in October/November 2008 preceded a sharp increase in illicit ivory trade. However, the correlation between this one-off sale and the increase in illegal ivory trade in subsequent years does not establish causality.

Although China and Japan were beneficiaries of the second one-off sale, these countries exhibit completely contrary patterns in the data. In China, involvement in illicit ivory trade has been growing incrementally every year since 1996, with the general annual rate of increase continuing without interruption after the one-off sale, to account for about 40% of all illegal ivory trade transactions in 2011. For Japan, involvement in illicit ivory trade has continued to decrease following the one-off sale and is considerably less than 1% in 2011. This strongly suggests that in both countries, other factors are operative that either exacerbate or mitigate ivory consumption beyond the impact of CITES one-off events.

Some of these factors are evident from the correlations that have been established through ETIS and MIKE research—for example, the degree of law enforcement, governance factors, consumer spending in end-use countries and so

d’éléphants natifs, tous ces acteurs sont des sources constantes d’ivoire illégal, et certains fonctionnent aussi comme un entrepôt ou un pays de transit et, à l’occasion, des points de sortie de l’ivoire qui quitte le continent africain. Taiwan est un lien potentiel du transit de l’ivoire qui passe à travers l’Asie, alors que l’Egypte abrite un grand marché intérieur d’ivoire non réglementé. Dans l’ensemble, les 9 pays et territoires prioritaires et les 10 pays et territoires secondaires de préoccupation sont collectivement responsables de deux tiers de toutes les activités du commerce illicite d’ivoire depuis la CdP15, selon l’analyse ETIS. La présence d’organisations du crime organisé est une caractéristique de mieux en mieux établie soutenant ce commerce illicite, et les lacunes de gouvernance portent gravement atteinte à l’application efficace de la loi tout au long de la chaîne commerciale.

L’analyse d’ETIS n’établit pas de corrélation entre la première vente d’ivoire exceptionnelle sous la CITES en juin 1999 et le commerce illicite ultérieur de l’ivoire. Après cet événement, la tendance de biais ajustée et lissée montre une augmentation progressive mais reste généralement plat dans son ensemble. Par contre, la seconde vente d’ivoire de la CITES en octobre/novembre 2008 a précédé une forte augmentation du commerce illicite de l’ivoire. Cependant, la corrélation entre cette vente exceptionnelle et l’augmentation du commerce illégal de l’ivoire dans les années suivantes n’établit pas de lien de causalité. Même si la Chine et le Japon ont été les bénéficiaires de la deuxième vente exceptionnelle, ces pays présentent des tendances totalement contraires dans les données. En Chine, l’implication dans le commerce illicite de l’ivoire a progressivement augmenté chaque année depuis 1996, et le taux général de croissance annuelle continue sans interruption après la vente exceptionnelle, pour constituer environ 40% de toutes les transactions du commerce illicite de l’ivoire en 2011. Pour le Japon, l’implication dans le commerce illicite de l’ivoire a continué à diminuer suite à la vente exceptionnelle et elle était nettement inférieure à 1% en 2011. Cela suggère fortement que, dans les deux pays, d’autres facteurs accentuent la consommation d’ivoire ou l’atténuent au-delà de l’impact des événements ponctuels de la CITES. Certains de ces facteurs sont évidents à partir des corrélations qui ont été établies par la recherche d’ETIS et de MIKE, par exemple, le degré d’application de la loi, les facteurs de gouvernance, les dépenses de consommation dans les pays d’utilisation finale etc., mais une évaluation exhaustive et une modélisation de tous les facteurs à l’origine de l’abattage illégal des éléphants et le commerce de l’ivoire est nécessaire pour établir les causes



on—but an exhaustive assessment and modelling of all drivers behind the illegal killing of elephants and trade in ivory is required to establish the causes or group of causes behind the increase. Such analysis constitutes a major undertaking that is beyond the scope of the ETIS and MIKE programmes at the present time.

At the same time, the ETIS report notes that in view of the escalating trend, there is little to suggest that implementation of the CITES action plan for the control of trade in African elephant ivory, adopted in 2005, has had measurable impact in curtailing illicit trade in ivory. Sanctions are overtly mentioned in the action plan of Decision 13.26, but they are almost never invoked, even against countries that are, as specified in the Decision, ‘found not to implement this action plan, or where significant quantities of ivory are found to be illegally sold’. Trade restrictions are also specified for lack of compliance with provisions for control of internal trade in Resolution Conf. 10.10 (Rev. CoP15). However, the only occasion in which sanctions were imposed was in 2008 against 13 elephant range States for an administrative infraction: failing to submit responses to the CITES Secretariat for an ivory trade questionnaire agreed at CITES CoP14. Countries that exhibit chronic failure to implement the long-standing requirements for internal trade in ivory as articulated in Resolution Conf. 10.10 (Rev. CoP15) or other actions called for in the annex to Decision 13.26 should be held accountable. Regrettably, the ETIS analyses seem to chronicle the same offenders time and time again. With a full-blown elephant crisis upon us, CoP16 provides the ideal opportunity for CITES parties to agree to stronger action.

TRAFFIC was able to present the ETIS analysis for CoP16 to government representatives from Asian elephant range States that attended the MIKE Sub-regional Steering Committee meeting for Southeast Asia held in Hanoi, Viet Nam, 3–4 December 2012. This meeting afforded an opportunity to review participation in ETIS and discuss ivory trade issues with a key Asian constituency, including representatives from Cambodia, China, Indonesia, Laos, Malaysia, Myanmar, Thailand and Viet Nam.

In other news, the three-year Darwin Initiative grant ‘Enhancing the Elephant Trade Information

ou groupes de causes à l’origine de l’augmentation. Une telle analyse constitue une entreprise majeure qui est au-delà de la portée des programmes d’ETIS et de MIKE à l’heure actuelle.

Dans le même temps, le rapport ETIS constate que, compte tenu de la tendance croissante, il y a peu de choses pour suggérer que la mise en œuvre de ce qu’on appelle « le plan d’action de la CITES pour le contrôle du commerce de l’ivoire de l’éléphant d’Afrique » adopté en 2005 a eu des impacts mesurables sur l’élimination du commerce illicite de l’ivoire. Des sanctions sont ouvertement mentionnées dans le « plan d’action » de la décision 13.26, mais elles ne sont presque jamais invoquées même contre des pays qui ne mettent pas en œuvre ce plan d’action, comme spécifié dans la décision, ou lorsqu’on trouve que des quantités importantes d’ivoire ont été vendues illégalement. Les restrictions au commerce sont également précisées pour manque de conformité avec les dispositions de contrôle du commerce intérieur dans la résolution Conf. 10.10 (Rev. CdP15). Cependant, la seule occasion où des sanctions ont été imposées était en 2008 contre 13 Etats de l’aire de distribution des éléphants pour une infraction administrative: d’avoir omis de soumettre leurs réponses au Secrétariat de la CITES pour un questionnaire sur le commerce de l’ivoire convenu à la CdP14. On devrait tenir responsables les pays qui présentent une incapacité chronique à mettre en œuvre les exigences de longue date pour le commerce international de l’ivoire comme stipulé dans la résolution Conf. 10.10 (Rev. CoP15) ou d’autres actions préconisées dans l’annexe à la décision 13.26. Malheureusement, les analyses d’ETIS semblent faire des rapports détaillés à maintes reprises sur les récidivistes. Etant donné la crise grave actuelle à laquelle les éléphants font face, la CdP16 offre une occasion idéale à la CITES de se mettre d’accord sur les mesures plus énergiques.

TRAFFIC a pu présenter l’analyse ETIS pour la CdP16 aux représentants des gouvernements des Etats de l’aire de distribution des éléphants d’Asie qui ont participé à la réunion du Comité directeur sous-régional pour l’Asie du Sud-Est qui s’est tenue à Hanoï au Viet Nam du 3 au 4 décembre 2012. Cette réunion était l’occasion d’examiner la participation à ETIS et de discuter les questions relatives au commerce de l’ivoire avec un groupe important d’Asie, y compris des représentants du Cambodge, de la Chine, de l’Indonésie, du Laos, de la Malaisie, de Myanmar, de la Thaïlande et du Viet Nam.

Concernant d’autres nouvelles, la subvention sur trois ans de l’Initiative de Darwin « Renforcement du Système d’Information sur le Trafic des Eléphants pour guider


ETIS update

System to guide CITES policy’ has now officially come to an end. This seminal project with the University of Reading places ETIS on the threshold of a promising new era through the delivery of a new ETIS online database system. As soon as a French version of the website is operational, the CITES Parties will have secure online access to all ivory seizure records that relate to their country at any time. Following a validation process under the auspices of the CITES Secretariat, designated representatives of each CITES party will be able to create new seizure records online, review and download their data, and access training materials and other resources whenever required. The online ETIS system is restricted to government data providers. As soon as both anglophone and francophone elephant range States can be equally serviced on the website, it will be launched with the CITES Parties. We expect to have this system up and running before the 16th meeting of the CITES Conference of Parties (CoP16) in early March 2013.

To support the long-term operational sustainability of ETIS, which has been dependent upon a core group of individuals since its inception, an ETIS Standard Operating Procedures Manual (SOP) has also been produced through the Darwin Initiative so that all of the procedures that relate to data collection, storage, retrieval, analysis and security are documented and institutionalized to ensure continuity into the future. While production of a SOP fills a major gap, we fully recognize that the manual is a living, iterative piece of work that will periodically be updated and augmented with further material as we move forward.

Another aspect of the Darwin Initiative project was to enhance the capabilities of national CITES management authorities to track and report elephant product seizure cases to ETIS. Toward that end, Jared Crawford has produced a new training module explaining the features and the functionality of the new online ETIS system to government data providers. This training programme will be available on the ETIS website for downloading and use as a self-tutorial whenever needed. With a series of detailed explanatory notes, we hope this offering will be a useful tool for assisting government authorities to make full use of the online ETIS facility.

la politique de la CITES » a officiellement pris fin. Ce projet clé fait avec l’Université de Reading place ETIS sur le seuil d’une nouvelle ère prometteuse grâce au un nouveau système de base de données d’ETIS en ligne. Dès qu’une version française du site sera opérationnelle, les Parties à la CITES auront un accès en ligne sécurisée à toutes les saisies d’ivoire qui se rapportent à leur pays à tout moment. Suite à un processus de validation sous les auspices du Secrétariat de la CITES, les représentants désignés de chaque Partie à la CITES seront en mesure de créer de nouveaux enregistrements de saisie en ligne, consulter et télécharger leurs données, et avoir accès au matériel de formation et d’autres ressources en cas de besoin. Le système en ligne d’ETIS est limité aux fournisseurs de données gouvernementales. Dès qu’on pourra servir de la même façon les Etats anglophones et francophones de l’aire de distribution des éléphants sur le site, on le lancera avec les Parties à la CITES. Nous nous attendons à ce que ce système soit en place et fonctionne avant la 16ème réunion de la Conférence des Parties de la CITES (CdP16) au début de mars 2013. Pour assurer la viabilité opérationnelle à long terme d’ETIS, qui dépend d’un petit groupe de personnes depuis sa création, un Manuel Standard des Procédures de Fonctionnement d’ETIS (MSPF) a également été réalisé grâce à l’Initiative Darwin pour que toutes les procédures qui se rapportent à la collecte de données, au stockage, à l’extraction, à l’analyse et à la sécurité soient documentés et institutionnalisés pour assurer la continuité dans le futur. Alors que la production d’un MSPF comble une lacune importante, nous sommes pleinement conscients que le manuel est un « ouvrage itératif vivant », qui sera régulièrement mis à jour et augmenté par d’autres documents au fur et à mesure que nous avançons.

Un autre aspect du projet de l’Initiative de Darwin était le renforcement des capacités des autorités nationales de gestion de la CITES de suivre et signaler les cas de saisies des produits issus de l’éléphant à ETIS. A cette fin, Jared Crawford a produit un nouveau module de formation expliquant les caractéristiques et la fonctionnalité du nouveau système en ligne d’ETIS aux fournisseurs de données gouvernementales. Ce programme de formation sera disponible sur le site Web d’ETIS pour le téléchargement et l’utilisation comme un support d’auto-apprentissage en cas de besoin. Avec une série de notes explicatives détaillées, nous espérons que cette offre sera un outil utile pour aider les pouvoirs publics à faire pleinement usage de la facilité en ligne d’ETIS.

Pour tester ce matériel et démontrer le fonctionnement du nouveau système en ligne, TRAFFIC a sponsorisé



To test these materials and demonstrate the new online system, under the Darwin Initiative grant TRAFFIC sponsored a two-day training workshop in Johannesburg, South Africa, on 17 and 18 September 2012 for 13 government authorities from Botswana, Ethiopia, Kenya, Malawi, Mozambique, Namibia, South Africa, Tanzania, Uganda, Zambia and Zimbabwe. All participants were given computers and internet access to develop the necessary skills for accessing and using the new online ETIS facility. Feedback from the event was overwhelmingly positive and participants were enthusiastic about the potential of accessing and using their ETIS data directly from the internet.

un atelier de formation de deux jours à Johannesburg, en Afrique du Sud les 17 et 18 septembre 2012, subventionné par l’Initiative Darwin, pour 13 autorités gouvernementales en provenance du Botswana, de l’Ethiopie, du Kenya, du Malawi, du Mozambique, de la Namibie, de l’Afrique du Sud, de la Tanzanie, de l’Ouganda, de la Zambie et du Zimbabwe. Tous les participants ont reçu des ordinateurs et un accès à l’Internet pour développer les compétences nécessaires pour accéder à et utiliser la nouvelle facilité en ligne d’ETIS. Des commentaires de l’événement étaient extrêmement positifs et les participants étaient enthousiastes concernant le potentiel de l’accès à et l’utilisation de leurs données d’ETIS directement à partir de l’Internet.


Guidelines for Contributors

Aim and scopePachyderm 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.

Submission of manuscriptsAll manuscripts should be submitted online at:http://pachydermjournal.org

If there are any questions or concerns regarding the submission process, please send an email to: [email protected] or otherwise contact by post/telephone:

The Editor, PachydermIUCN/SSC AfESGPO Box 68200 – 00200Nairobi, Kenyatelephone: +254 20 249 3561/65fax: +254 20 2493570

Manuscripts are accepted in both the English and French languages. Where possible, the abstract should be provided in both languages.

Pachyderm’s Editorial Board categorizes material received into the following sections:

Research and management papersThese should be not more than 5,000 words and be structures as follows: 1) Title, 2) Abstract of not more than 250 words (informative type, outlining information from the Introduction, Materials and

GUIDELINES FOR CONTRIBUTORS

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 (no more than 25), 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.

Field notesThe journal welcomes notes from the field. They may contain figures and tables but should be < 2,500 words.

Review papersReview papers, which are unbiased reviews of all the existing knowledge on a specific topic, are welcomed. Length should be < 6,000 words

Book reviews

Pachyderm invites reviews of newly published books, which should be < 1500 words.

Letters to the editorLetters should be addressed to the relevant Specialist Group Chair, and should be < 1,000 words. Letters are welcome that comment on articles published in Pachyderm or on any other issue relating to elephant and rhino conservation in the wild.

Preparation of manuscripts

Images, figures and mapsPreferably provide figures and maps in their original form, for example, charts and data in Excel files, maps as EPS and images in the highest quality possible, such as TIF (600 dpi). Indicate clearly the author or source of figures, maps and photographs.


Guidelines for contributors

Title and authorsThe 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 institutional affiliation and full postal and email address of the corresponding author, to whom proofs and editorial comments will be sent.

Journal conventions

NomenclatureUse common names of animals and plants, giving scientific names in italics on first mention. Generally refer to animals in the plural form (i.e. rhinos, elephants).

SpellingUse British spelling, following the latest edition of the Concise Oxford dictionary or the Oxford English Dictionary, using ‘z’ instead of ‘s’ in words like ‘recognize’, ‘organization’, ‘immobilized’; but ‘analyse’, ‘paralyse’. The dictionary is available online at http://oed.com.

NumbersUse the International System of 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, use a comma as the separator for figures four digits or more: 1,750 and 11,750. The separator will be a full stop in French papers.Numbers

ReferencesWe use the name-year method of citing and listing references. As of the next issue, the punctuation and typographic style we will use are as advocated by the internationally recognized Council of Science Editors in its Scientific style and format, 7th edition.

In the text, cite a single author: ‘(X 2005) or ‘X (2005; cite two authors: ‘(X and Y 2005)’ or ‘X and Y (2005)’; cite more than two authors ‘(X et al. 2007)’ or ‘X et al. (2007)’. Note that there is no comma between the author(s) and the year. If multiple works are being cited, separate them by a semi-colon, listing them in chronological order: (X et al. 1998; B 2002; Z 2010).

In the reference list, punctuation is minimized. Examples are drawn from previous issues of Pachyderm:

Article in a journal or periodicalBarnes RFW, Barnes KI, Alers MPT, Blom A. 1991.

Man determines the distribution of elephants in the rain forests of northeastern Gabon. African Journal of Ecology 29:54–63.

Book

White I, Edwards A, eds. 2000. Conservation research in the African rain forests: a technical handbook. Wildlife Conservation Society, New York.

Smithers RHN.1983. Mammals of the southern African sub-region. 2nd ed. Pretoria University, Pretoria.

Chapter in a book

Barnes RFW. 1996. Estimating forest elephant abundance by dung counts. In: Kangwana K, ed. Studying elephants. AWF Technical Handbook no. 7. African Wildlife Foundation, Nairobi. p. 33–48.

Unpublished material

Adcock K. 2006. Estimates of black rhino carrying capacity at Ol Pejeta’s new rhino area. Kenya Wildlife Service, Nairobi. Unpublished.

Blake S 2002. The ecology of forest elephant distribution and its implications for conservation. PhD thesis. Institute of Cell, Animal and Population Biology, University of Edinburgh, Edinburgh.

Electronic siteElephants of Cameroon. 2000. Saving Africa’s

vanishing giants, the elephants of Cameroon.


Guidelines for Contributors

http://www.nczooeletrack.org/project/index.htm. Accessed 25 February 2000.

[AfESG] African Elephant Specialist Group. 2000. Fencing and other barriers against problem elephants. AfESG Technical Brief Series. IUCN African Elephant Specialist Group, Human–Elephant Conflict Working Group (author: Richard Hoare). Available at: http://www.african-elephant.org/hec/pdfs/hecfencen.pdf.

journal of the african elephant, african rhino and asian rhino

Documents