KWAME NKRUMAH UNIVERSITY OF SCIENCE AND TECHNOLOGY COLLEGE OF AGRICULTURE AND NATURAL RESOURCES FACULTY OF RENEWABLE NATURAL RESOURCES DEPARTMENT OF WILDLIFE AND RANGE MANAGEMENT THE ECOLOGY AND MANAGEMENT OF ELEPHANTS (Loxodonta africana Matschie) IN THE BIA – GOASO FOREST ENCLAVE OF WESTERN GHANA By MOSES KOFI SAM B.Sc. NRM (Hons), M.Sc. Conservation Biology A thesis submitted to the Department of Wildlife and Range Management Kwame Nkrumah University of Science and Technology in partial fulfillment of the requirement for the degree of DOCTOR OF PHILOSOPHY Faculty of Renewable Natural Resources College of Agriculture and Renewable Natural Resources October 2010 DECLARATION I hereby declare that this submission is my own work towards the PhD and that, to the best of my knowledge, it contains no material previously published by another person nor material which has been accepted for the award of any other degree of the University, except where due acknowledgement has been made in the text. ………………………… …………………………… …………………………… Student‟s Name & ID Signature Date Certified by: ………………………… …………………………… …………………………… Supervisor‟s Name Signature Date Certified by: ………………………… …………………………… …………………………… Head of Dept. Name Signature Date i ABSTRACT Elephants are important mega fauna whose role in an ecosystem could ensure the health and survival of diverse species. However, drastic declines in their numbers since the 19th Century in Africa in general and West Africa in particular has necessitated the need to understand their ecology in the local context to help improve their survival. This study was therefore conducted in the Bia-Goaso forest enclave in Ghana to contribute to the achievement of the objectives of the Elephant Conservation Strategy of Ghana (2000) and the “Action plan for the Management of the Transfrontier Elephant Conservation Corridor of Bia-Goaso-Djambarakrou,” which contain activities geared towards ensuring the survival of forest elephants in Ghana and West Africa respectively. In February 2004, a dry season survey on elephants and an investigation into the different levels of human and ecological variables that affect their abundance and distribution was conducted in an extensive network of eleven forest reserves and two wildlife reserves in the Bia–Goaso enclave of western Ghana. These activities were repeated twice, first in the rainy season of 2007 and the following dry season which started in 2007 and ended in 2008 mainly to understand seasonal effects in distribution of elephants in the area. During this period, the most up-to-date method for surveying forest elephants, the retrospective method was employed. A desk-top exercise was then undertaken not only to prove the importance or otherwise of the elephant populations in the study area in comparison with other elephant populations in West Africa but also to determine factors that govern their abundance. A preliminary investigation into the feasibility of creating corridors between reserves in the study area and neighbouring Ivory Coast was initiated in May 2004. As part of this, the movements of elephants from Ghana into Ivory Coast along the Bia river was monitored every quarter between August 2005 and December 2006, as well as an investigation into the rate of forest degradation. The nature and extent of human – elephant conflicts in the Bia Conservation Area was also studied through administration of questionnaires using interviews and field measurements. One hundred and thirty (130) transects were systematically distributed in three strata (high, medium and low density) based on dung pile density estimates in an initial reconnaissance. Two models (rainfall and steady state assumption models) were used to estimate elephant dung pile density and numbers in the study area. However, the rainfall model is preferred since it ii makes no assumption about the state of the forest. Two major elephant populations were observed to be residing in the study area, with the possibility of a tiny third population of less than 10 elephants (in the Bia North FR). In the 2004 survey, a mean population estimate of 115 (CL: 90 - 148) elephants was obtained for the Bia Resource Reserve. The eastern portion of the Mpameso Forest Reserve (medium density stratum) had an estimate of 57 (CL: 33 - 100) elephants. Elephant numbers could not be estimated for the rest of the reserves (low-density stratum) because of inadequate number of dung piles on transects. In the 2007/8 survey, estimates of 133 (CL: 104 - 162) elephants and 137 elephants (CL: 98 - 170) elephants were obtained for Bia CA during the dry and wet season surveys respectively. While for the Mpameso elephant range, estimates of 83 (CL: 41 - 125) elephants and 90 (CL: 49 - 131) elephants were obtained in the dry and wet seasons respectively. Merged estimates for both seasons were 135 (CL: 114 - 156) elephants for Bia and 87 (CL: 58 - 116) elephants for Mpameso. Altogether, a total population of 172 (CL: 123- 264) elephants were estimated to be occurring in the entire study area. Elephants were found to be clumped more to the southeast of BRR, the eastern part of Mpameso and the Bia Shelterbelt FR. Elephants were found to be more widely spread out in the wet season than in the dry season within BCA. Analysis of dung pile distribution in relation with human and ecological variables in both the 2004 and 2007/8 studies showed that within the reserves, water availability explained a high proportion (ca. 90%) of the variance in elephant density, with elephants being clumped around watering points created as a result of logging. Additionally, in 2004, distance to the Bia River was inversely correlated to the number of dung piles seen per km in the Mpameso Forest Reserve. Illegal activities (such as snaring and snail picking), however, did not affect elephant abundance but rather had a negative correlation with watering points, that is, they were undertaken away from these watering points. This suggests that poachers were avoiding areas of high water availability, possibly because of high elephant activity around those areas. Outside reserves, the distance to major towns and roads accounted largely for variances in elephant density in 2004. In 2007/8 on the other hand, logging roads and availability of raphia also entered the model as significant variables, though they did not add much power to the models. By all standards, the Bia population on its own is a very important one in the sub-region. The Goaso population is smaller. However, the security of its range, and the possibility of linking it with others makes it also even iii more important for the overall conservation of forest elephants in West Africa. It was established during the study that elephant abundance in West Africa depends on the size of a population‟s range, to whether the range is protected, its geology and the quality of governance in the country where it occurs. Elephant crop damage is a serious problem in the area, especially to cluster of farms that border the southern portions of the Bia Resource Reserve, resulting in conflicts between farmers and nature conservation. Food crops such as plantain, cassava, cocoa and maize suffered severest damage. These are most preferred by the elephants when in the matured state and the quality is excellent. For each community, that suffered crop-raiding, the extent of crop-raiding depended not only on the area of land under cultivation but also the mean distance of farms from the reserve boundary, as well as the number of different crops planted on the farm. Interviews conducted in 2004 indicated that most migrant farmers do not see any advantage in preserving elephants and would do little to conserve them. While they were not willing to sacrifice part of their already overburdened farmlands for establishing elephant corridors, they also feared a possible increase in human-elephant conflicts in the future. However, most villages adjacent to streams/rivers showed a strong interest in restoring the riparian forest since they faced water and fish shortages in the dry season. More lately, there have been several forest restorative activities going on in the study area in the hope of returning wildlife into communal lands. Hence, there has been a generally increasing level of awareness concerning the conservation and protection of lands to serve as corridors. Of six potential corridors considered, one along the Bia river linking the Goaso population to those in Ivory Coast and from Bia Resource Reserve to Krokosue Hills Forest Reserve have the greatest potential in the short to medium term, and it is also likely that with a high level of interventions, two others could be achieved over the long term. A more detailed spatial analysis in combination with ground truthing is required before a conclusive decision on viable corridors can be made. Also a more participatory information gathering strategy (such as Timelines, H-Diagram) for soliciting community opinions on corridors have been suggested to complement the use of questionnaires. Recommendations also include a detailed study of elephant movements in the Ivorian side of the corridor. iv ACRONYMS AfESG African Elephants Specialists Group BCA Bia Conservation Area BP British Petroleum CA Conservation Area CREMA Community Resource Management Area DGW Department of Game and Wildlife EU European Union FC Forestry Commission FR Forest Reserve FRNR Faculty of Renewable Natural Resource GPS Global Positioning System HECs Human- elephant‟s conflicts IUCN International Union for Conservation of Nature KNUST Kwame Nkrumah University of Science and Technology MTC Mim Timber Company NP National Park NTFP Non Timber Forest Product OSS Optimum Sample Size PAC Problem Animal Control PADP Protected Areas Development Programme PAs Protected Areas RR Resource Reserve SB Shelter Belt SSAM Steady State Assumption Model SSC Species Survival Commission WD Wildlife Division WWF World Wide Fund for Nature v TABLE OF CONTENTS PAGES DECLARATION I ABSTRACT II ACRONYMS V TABLE OF CONTENTS VI LIST OF FIGURES XI LIST OF TABLES XIV ACKNOWLEDGEMENTS XVII CHAPTER 1.0 INTRODUCTION 1 1.1 Justification for this study 1 1.2 Objectives of study 3 CHAPTER 2.0 LITERATURE REVIEW 4 2.1 Description of Study Area 4 2.1.1 Location and Administrative Divisions 4 2.1.2 Socio-economic profile 5 2.1.3 Climate 6 2.1.4 Topography, soils and drainage 6 2.1.5 Flora 7 2.1.6 Fauna 9 2.2 Distribution and status of Elephant 11 2.2.1Distribution and status of elephant in Africa 11 2.2.2 Distribution and status of elephants in West Africa 12 2.2.3 Distribution and status of elephants in Ghana 13 2.2.4 Conservation Importance of the Bia –Goaso elephant populations 16 2.3 Factors that influence elephant distribution and numbers 17 2.4 Decline in Elephant Populations 20 2.4.1 Decline in Elephant population in Africa 20 2.4.2 Decline in Elephant population in West Africa 22 2.4.3 Decline in Elephant population in Ghana 23 2.5 Habitat Fragmentation 24 2.5.1 Fragmentation of Elephant in Ghana 25 2.6 Habitat corridors 26 2.6.1 Rationale for creating Habitat Corridors 27 2.6.2 Conservation Importance of Habitat Corridors for Elephants and other animals 27 2.6.3 Relevance of Habitat Corridors in West Africa 29 2.6.4 Relevance of Trans-boundary Ivory Coast-Ghana Corridor 30 2.6.5 Characteristics of Successful Corridors 31 2.7 Human-elephant conflicts 33 vi 2.7.1 Factors influencing crop-raiding 34 2.7.2 Mitigating crop-raiding 37 CHAPTER 3.0 DISTRIBUTION AND NUMBERS OF ELEPHANTS IN BIA-GOASO FOREST ENCLAVE 40 3.1 Introduction 40 3.2 Research objectives and questions 41 3.3 Hypotheses 42 3.4 Methods 42 3.4.1 Reconnaissance Survey 42 3.4.2 Field Survey during the dry season of 2004 year 43 3.4.3 Field Survey during wet and dry seasons of 2007/8 years 47 3.5 Data Analysis 50 3.5.1 Calculating elephant dung-pile density using the Steady State Model 51 3.5.2 Calculating number of elephants using the Rainfall Model 52 3.5.3 Estimation of elephant using the retrospective method 53 3.5.4 Elephant distribution 54 3.6 Results 55 3.6.1 Estimates of elephant numbers in the 2004 dry season 55 3.6.2 Estimates of elephant numbers in 2007/8 60 3.6.3 Estimation of dung decay rate for 2007/8 65 3.6.4 Estimation of elephant numbers for 2007/8 66 3.6.5 Distribution of elephants in the study area in 2004 68 3.6.6 Distribution of elephants in the study area in 2007/8 72 3.7 Discussion 72 3.7.1 Numbers of elephants in BCA 72 3.7.2 Distribution of elephants 75 3.8 Conclusions 75 CHAPTER 4.0 RELATIONSHIP BETWEEN ELEPHANT OCCUPANCY AND HUMAN AND ECOLOGICAL VARIABLES IN THE BIA-GOASO FOREST ENCLAVE 77 4.1 Introduction 77 4.2 Research objectives and questions 78 4.3 Hypotheses 78 4.4 Methods 79 4.4.1 Habitat evaluation 79 4.4.2 Factors affecting elephant abundance within reserves 79 4.4.3 Factors affecting elephant abundance outside reserves 80 4.5 Data Analysis 81 4.6 Results 82 4.6.1 Factors affecting elephant abundance within study reserves during the 2004 study period 82 4.6.2 Factors affecting elephant abundance outside reserves (2004) 87 4.6.3 Factors affecting elephant distribution (outside/inside) reserves during the 2007/8 study period 89 4.6.4 Illegal activity 91 4.7 Discussion 91 4.7.1 Factors within Reserves 91 4.7.2 Factors outside Reserves 96 vii 4.8 Conclusion 96 CHAPTER 5.0 POTENTIAL CORRIDORS FOR ELEPHANT MOVEMENT BETWEEN FORESTS RESERVES IN THE BIA-GOASO FOREST ENCLAVE AND THE BORDERING FORESTS IN IVORY COAST 97 5.1 Introduction 97 5.2 Research objectives and questions 99 5.3 Hypotheses 100 5.4 Methods 100 5.4.1 Elephant movement along the Bia River 100 5.4.2 Change detection using Post classification 102 5.4.3 Using satellite imagery to determine spatial and biological feasibility for creating corridors 103 5.4.4 Questionnaire survey to determine socio-economic feasibility 5.4.5 A reconnaissance trip to the Ivory Coast side of the border adjacent to the study area 104 5.5 Data Analysis 105 5.5.1 Elephant movement along the Bia River 105 5.5.2 Geographical/ecological feasibility 105 5.5.3 Socio-economic feasibility 106 5.6 Results 107 5.6.1 Movement of elephants along the Bia River 107 5.6.2 Land cover changes from 1989 to 2003 107 5.6.3 Geographical/biological feasibility of corridors 109 5.6.4 Socio-economic feasibility of corridors 110 5.7 Discussion 113 5.7.1 Human dimension considerations in corridor creation 113 5.7.2 Biological and spatial considerations in corridor creation 115 5.8 Conclusion 120 CHAPTER 6.0 HUMAN-ELEPHANT CONFLICTS SITUATION AROUND THE BIA CONSERVATION AREA (BCA) 122 6.1 Introduction 122 6.2 Research objectives and questions 122 6.3 Hypotheses 123 6.4 Methods 124 6.4.1 Social survey of human-elephant conflicts 124 6.4.2 Field assessment of elephant damages 124 6.4.3 Determining elephants involved in crop-raiding 125 6.4.4 Comparing dung-pile density immediately inside and outside the BCA 125 6.5 Data Analysis 126 6.6 Results 127 6.6.1 Severity of crop raiding 127 6.6.2 Elephants involved in crop raiding 133 6.6.3 Elephant distribution versus crop-raiding 135 6.7 Discussion 137 6.7.1 Nature and extent of HECs 137 6.7.2 Elephants involved in crop-raiding and their movement around boundary of BCA 141 6.8 Conclusion 142 viii CHAPTER 7.0 Comparative analysis of the conservation status of the Bia-Goaso elephants with other forest elephant populations in West Africa 7.1 Introduction 144 7.2 Research objectives and questions 147 7.3 Hypotheses 148 7.4 Methods 148 7.4.1 Prioritising elephant populations in West Africa 148 7.4.2 Governance Indicators and other variables that determines elephant abundance in West Africa 149 7.5 Data Analysis 151 7.5.1 Prioritisation of elephant populations 151 7.5.2 Governance indicators that determines elephant numbers/densities 152 7.6 Results 154 7.6.1 Importance of the Bia-Goaso elephant population and their range for the conservation of elephants in West Africa 154 7.6.2 Indices of governance and other variables that determine elephant abundance in West Africa 158 7.7 Discussion 164 7.7.1 Importance of Area for Elephant Conservation in West Africa 164 7.8 Conclusion 166 CHAPTER 8.0 GENERAL DISCUSSION 167 8.1 Status of elephants in the Bia-Goaso forest enclave 167 8.2 Pachyderms, Plantains and Poverty 168 8.2.1 Pachyderms - profile and behavior of elephants involved in crop-raiding 170 8.2.2 Plantains - the case of cultivated plants palatable to elephants 171 8.2.3 Degree of poverty in communities resulting from elephant crop- raiding activities 172 8.3 Conservation, Corridors, and CREMAS 174 8.3.1 Conservation and long term survival of elephants of the Bia-Goaso forest enclave 174 8.3.2 Corridors within the Bia-Goaso forest enclave 176 CHAPTER 9.0 GENERAL CONCLUSIONS AND RECOMMENDATIONS 9.1 Conclusions 185 9.2 Recommendations 186 CHAPTER 10.0 REFERENCES 190 APPENDICES APPENDIX 1: Survey of attitudes of members of fringe communities in selected forest reserves in the Bia-Goaso forest enclave on issues of corridor creation and elephant management 201 APPENDIX 2: Activities of elephants on transects along the Bia River ix