The University of Dodoma University of Dodoma Institutional Repository http://repository.udom.ac.tz

Business Doctoral Theses

2020 Contribution of Bushmeat to household food and income and factors influencing household dependence on Bushmeat in western Serengeti

Manyama, Flora Felix

The University of Dodoma

Manyama, F. F. (2020). Contribution of Bushmeat to household food and income and factors influencing household dependence on Bushmeat in western Serengeti (Doctoral thesis). The University of Dodoma, Dodoma. http://hdl.handle.net/20.500.12661/2416 Downloaded from UDOM Institutional Repository at The University of Dodoma, an open access institutional repository. CONTRIBUTION OF BUSHMEAT TO HOUSEHOLD FOOD AND INCOME AND FACTORS INFLUENCING HOUSEHOLD DEPENDENCE ON BUSHMEAT IN WESTERN SERENGETI

FLORA FELIX MANYAMA

DOCTOR OF PHILOSOPHY THE UNIVERSITY OF DODOMA OCTOBER, 2020 CONTRIBUTION OF BUSHMEAT TO HOUSEHOLD FOOD AND INCOME AND FACTORS INFLUENCING HOUSEHOLD DEPENDENCE ON BUSHMEAT IN WESTERN SERENGETI

BY FLORA FELIX MANYAMA

A DISSERTATION SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

THE UNIVERSITY OF DODOMA OCTOBER, 2020 DECLARATION

AND

COPYRIGHT

I, Flora Felix Manyama, declare that this dissertation is my own original work and that it has not been presented and will not be presented to any other university for a similar or any other degree award. All the sources of materials used to accomplish this work are duly acknowledged.

Signature:

No part of this dissertation may be reproduced, stored in any retrieval system, or transmitted in any form or by any means without prior written permission of the author or the University of Dodoma. If transformed for publication in any other format shall be acknowledged that, this work has been submitted for degree award at the University of Dodoma.

i CERTIFICATION

The undersigned certifies that they have read and hereby recommends for acceptance by the University of Dodoma, a dissertation entitled “Contribution of Bushmeat to Household Food and Income and Factors influencing household dependence on bushmeat in Western Serengeti” in partial fulfilment of the requirements of the degree of Doctor of Philosophy in Environmental Science and Conservation of the University of Dodoma.

Professor Julius W. Nyahongo

Signature Date (FIRST SUPERVISOR)

Professor Eivin Røskaft

Signature Date 12/06/2020 (SECOND SUPERVISOR)

ii DEDICATION

This work is dedicated to my beloved parents Felix Manyama and Loyce Kasika for their guidance and support in my life. Many thanks for your encouragement throughout my life and academic journey.

iii ACKNOWLEDGEMENTS

I am grateful for the number of people who made this work accomplished. Special thanks go to the almighty God for his blessings throughout my PhD study which made the work successfully. I express my special thanks and appreciation for the learning opportunity provided by my Employer the University of Dodoma (UDOM). Thanks are also extended to AfricanBioServices Project and UDOM for financial support in this study.

I would like to express my sincere gratitude to my supervisors Professor Julius W. Nyahongo from UDOM, Professor Eivin Røskaft from the Norwegian University of Science and Technology (NTNU), Norway and Professor Martin R. Nielsen from the University of Copenhagen, Denmark for their guidance and support during the entire period of my study.

My heartfelt appreciation is also extended to Professor Jonathan Kabigumila, Professor Abiud Kaswamila and Dr. Chrispinus Rubanza for their support to make this work accomplished. Moreover, I would like to express my deepest thanks to government institutions; TAWIRI, COSTECH and NIMR for granting me research permit to undertake this study.

I would like to acknowledge with gratitude the support of district officers in Serengeti, Rorya and Butiama districts for technical support and field assistance. I am also thankful to village leaders of Robanda, Rwamkoma and Kowak for their cooperation during research work.

Special thanks go to research assistants; Patrick Semwenda, Havijawa Bilali, Baraka Majugwa, Kimani Magori, Deogratius Makuru, George Nyahongo, and primary school teachers for their support during field work in their areas. I also thank all the respondents in the villages where this study was conducted.

Thanks are also extended to the Department of Biology at UDOM for hosting me during the entire period of my study. I also appreciate the support from UDOM staff and my fellow PhD students; Moses T. Kyando, Agnes Kisanga, Doreen Mrimi and Sr. Edtruda Mbegu for their cooperation during our studies.

iv I acknowledge the support from AfricanBioServices research team and TAWIRI staff for their cooperation and technical support. Thanks to Dr. Peter S. Ranke from NTNU for R-tutorials which has contributed a lot in data analysis for this work. Thanks to Dr. Kwaslema Malle Hariohay from TAWIRI for his support in drawing maps for this study and all the people from the project for their contribution to make the work successful.

I am extremely grateful to my parents for their love, prayers, caring and sacrifices for educating and preparing me for my future. I am very much thankful to my family and my daughter Careen for their love, understanding, prayers and continuing support to complete my studies.

Finally, I would like to thank all the people who have supported me in one way or another and contributed to the success of this study, thank you so much may God bless you.

v ABSTRACT

Bushmeat is an important source of household (HH) food and income in western Serengeti although information on the frequency of consumption and income earned is unknown due to the illegal nature of the activity. This study was intended to determine the contribution of bushmeat to HHs and factors influencing bushmeat dependence. The study was conducted in three villages (Robanda, Rwamkoma and Kowak) selected purposely based on distances from the western boundary of Serengeti National Park (SNP). Data were obtained through HH questionnaire surveys, dietary recall surveys and observations and recording of bushmeat packages conducted in both the dry (September-October 2017) and wet (April-May 2018) seasons. Data on bushmeat consumption frequencies were collected from 127 schoolchildren and compared that to 150 adults from regular HHs selected randomly. Also snowballing was used to locate hunters and bushmeat traders where, 96 respondents were identified.

Overall, bushmeat contribute by 15.8% of all meat sources reported but its contribution was more in the closest village (96.3%), declining with distance from SNP (Kruskal-Wallis test; H=454.2; P<0.001). Bushmeat was consumed more frequently during the dry season (66%) compared to the wet season (34%). Adults on average reported significantly lower bushmeat consumption frequencies than schoolchildren (Wilcoxon test; W=33,526; P=0.003) which imply that children can provide reliable information about the importance of bushmeat in HH consumption than adults. The generalised linear model revealed that, bushmeat consumption in HH was significantly influenced by season, distance and consumption of other meat sources (Table 4.3). The contribution of bushmeat to HH income was significantly higher in the closest village than in the intermediate and distant villages (Kruskal-Wallis test; H=24.025; P<0.001). HH reliance on bushmeat income was negatively associated with age and gender of the HH head and distance to the protected area (PA) boundary. Hence, efforts to reduce illegal hunting should target male-headed HHs close to PA boundary through promoting alternative meat and income sources.

vi TABLE OF CONTENTS

DECLARATION AND COPYRIGHT ...... i CERTIFICATION ...... ii DEDICATION ...... iii ACKNOWLEDGEMENTS ...... iv ABSTRACT ...... vi TABLE OF CONTENTS ...... vii LIST OF TABLES ...... xi LIST OF FIGURES ...... xii LIST OF PLATES ...... xiv LIST OF APPENDICES ...... xv LIST OF ABBREVIATIONS AND ACRONYMS ...... xvi

CHAPTER ONE ...... 1 INTRODUCTION ...... 1 1.1 Background of the study ...... 1 1.2 Statement of the Research Problem ...... 4 1.3 Objectives of the Study ...... 5 1.3.1 Specific Objectives...... 5 1.4 Hypotheses ...... 6 1.5 Significance of the Study ...... 6 1.6 Organization of the Dessertation ...... 7

CHAPTER TWO ...... 9 LITERATURE REVIEW ...... 9 2.1 Overview ...... 9 2.2 Definition of Key Terms ...... 9 2.2.1 Bushmeat Hunting ...... 9 2.2.2 Protected Areas ...... 9 2.2.3 Ecosystem Goods and Services ...... 9 2.2.4 Local Communities ...... 10 2.3 Theoretical Review ...... 10 2.3.1 The Social Exchange Theory ...... 10 vii 2.3.2 The Rational Choice Theory ...... 11 2.4 Empirical Review ...... 11 2.4.1 Tanzania Wildlife Resources Management Policy Frameworks ...... 12 2.4.1.1 The National Environmental Policy ...... 12 2.4.1.2 The National Wildlife Policy ...... 13 2.4.1.3 The National Forest Policy ...... 14 2.4.1.4 The National Tourism Policy ...... 15 2.4.1.5 The National Land Policy ...... 16 2.4.2 Impelementation of International Wildlife Policy Framewok in Tanzania ...... 16 2.4.2.1 The Convention on Biological Diversity ...... 17 2.4.2.2 Convention on International Trade in Endangered Species of Wild Fauna and Flora ...... 17 2.4.3 Wildlife Resources Utilization in Tanzania ...... 18 2.4.3.1 Status and extent of Bushmeat Hunting in Tanzania ...... 18 2.4.4 The Importance of Bushmeat ...... 19 2.4.5 Bushmeat Hunting in Western Serengeti ...... 20 2.4.6 The Bushmeat Market Structure and Trading System ...... 21 2.4.7 Factors Influencing Bushmeat Hunting ...... 23 2.4.8 Approaches in Studying Illegal Bushmeat Hunting ...... 24 2.5 Knowledge Gap ...... 25 2.6 Conceptual Framework ...... 26

CHAPTER THREE ...... 28 METHODOLOGY ...... 28 3.1 Overview ...... 28 3.2 Description of Study Area ...... 28 3.3 Description of Study Villages ...... 30 3.3.1 Robanda Village ...... 31 3.3.2 Rwamkoma Village ...... 32 3.3.3 Kowak Village ...... 32 3.4 Research Design and Sampling ...... 33 3.4.1 Sampling Unit ...... 33 3.4.2 Sample Size ...... 33

viii 3.4.3 Sampling procedures ...... 33 3.5 Data Collection Methods ...... 34 3.5.1 Household Questionnaire Survey ...... 37 3.5.2 Dietary Recall Survey ...... 37 3.5.3 Focused Group Discussion ...... 38 3.5.4 Observations and Recording of Bushmeat Packages ...... 38 3.6 Data Collection...... 39 3.6.1 Data Collection from Schoolchildren...... 39 3.6.2 Data Collection from Adults ...... 41 3.7 Data Analysis ...... 43 3.8 Validity and Reliability ...... 49 3.8.1 Validity ...... 49 3.8.2 Reliability ...... 49 3.8.3 Ethical Considerations ...... 49 3.9 Delimitations and Limitations of the Study ...... 50 3.9.1 Delimitations of the Study ...... 50 3.9.2 Limitations of the Study ...... 50

CHAPTER FOUR ...... 52 RESULTS ...... 52 4.1 Overview ...... 52 4.2 Demographic and socioeconomic characteristics of the respondents ...... 52 4.3 Seasonal variation in frequency of bushmeat consumption by households ...... 55 4.4 Contribution of bushmeat to household meat consumption as a function of distance from the SNP boundary ...... 56 4.4.1 Contribution of other meat types to household meat consumption as a function of distance from the SNP boundary ...... 57 4.5 Variation in the frequency of bushmeat consumption reported by adults and Schoolchildren ...... 59 4.5.1 Variations in the consumption frequency of other meat types reported by adults and schoolchildren ...... 61 4.6 Factors influencing bushmeat consumption in households ...... 62 4.7 Demographic and socio-economic characteristics of adult respondents ...... 63

ix 4.8 Contribution of bushmeat to household income and the influence of distance from PA boundary on bushmeat hunting activities ...... 65 4.8.1 Bushmeat hunting activities and bushmeat packages observed and recorded ...... 67 4.8.2 Contribution of other income sources to total household income as a function of distance from PA boundary ...... 71 4.9 Factors influencing household participation in hunting and trading bushmeat and household bushmeat income reliance ...... 76

CHAPTER FIVE ...... 78 DISCUSSION ...... 78 5.1 Overview ...... 78 5.2 Hypothesis 1: Bushmeat consumption frequency is higher during the dry season than the wet season ...... 78 5.3 Hypothesis 2: Bushmeat consumption frequency decreases with increasing distance from protected area boundary ...... 79 5.4 Hypothesis 3: Schoolchildren report higher bushmeat consumption frequency than adults ...... 80 5.5 Hypothesis 4: Bushmeat consumption frequency is associated with household socioeconomic and other characteristics ...... 80 5.6 Hypothesis 5: Bushmeat hunting and household reliance on bushmeat income decreases with distance from PA boundary ...... 81 5.6.1 Source of bushmeat ...... 83 5.7 Hypothesis 6: Socioeconomic factors influencing household participation in hunting and reliance on bushmeat income ...... 84

CHAPTER SIX ...... 87 CONCLUSION AND RECOMMENDATIONS ...... 87 6.1 Conclusion ...... 87 6.2 Recommendations ...... 88 6.2.1 Recommendations for further research ...... 89 REFERENCES ...... 91 APPENDICES ...... 107

x LIST OF TABLES

Table 4. 1: Baseline information of the respondents (schoolchildren; n = 127) ...... 53 Table 4. 2: Baseline information of the adult respondents from regular households ...... 54 Table 4. 3: Regression coefficients of the quasi-Poisson Generalised Linear Model for predictors of household bushmeat consumption frequency ...... 63 Table 4. 4: Percentage composition of the adult sample and comparison of means between hunters/bushmeat traders and regular households ...... 64 Table 4. 5: Average bushmeat price per 1kg of different wildlife species (1 US$ = TZS 2,250) ...... 71 Table 4. 6: Test of significance differences in mean income and reliance on income from different sources for hunter and bushmeat trader and regular households, overall and in each village...... 72 Table 4. 7: Heckman sample selection model predicting household participation in hunting and trading bushmeat and reliance on bushmeat income...... 77

xi LIST OF FIGURES

Figure 2. 1: A Conceptual framework describing the role of bushmeat in households and factors influencing household consumption and bushmeat income reliance in western Serengeti...... 27 Figure 3. 1: Map of the study area and its location in Tanzania (left) showing the study villages Robanda, Rwamkoma and Kowak indicated with black dots...... 30 Figure 4. 1: Average number of bushmeat meals consumed per week reported by adults and schoolchildren in the wet and the dry season in Robanda, Rwamkoma and Kowak at increasing distance from the PA boundary...... 56 Figure 4. 2: Average number of domestic meat meals consumed per week reported by adults and schoolchildren in the wet and dry season in Robanda, Rwamkoma and Kowak at increasing distance from the PA boundary...... 58 Figure 4. 3: Average number of fish meals consumed per week reported by adults and schoolchildren in the wet and dry season in Robanda, Rwamkoma and Kowak at increasing distance from the PA boundary...... 59 Figure 4. 4: Average bushmeat, livestock, crop and wage and business income of hunter and bushmeat trader households at increasing distance from the PA boundary...... 65 Figure 4. 5: Reliance (percentage contribution to total household income) of bushmeat, livestock, crop and wage and business income of hunter and bushmeat trader households at increasing distance from the PA boundary...... 66 Figure 4. 6: Wildlife species contribution to total bushmeat packages recorded in the dry and the wet season...... 67 Figure 4. 7: Map the study area and its location in Tanzania (left) showing the study villages (Robanda, Rwamkoma and Kowak) and the bushmeat market centers indicated with red dots...... 70 Figure 4. 8: Average total income in hunter and bushmeat trader and regular households at increasing distance from the PA boundary...... 72 xii Figure 4. 9: Average crop, livestock and wage and business income of regular households at increasing distance from the PA boundary...... 74 Figure 4. 10: Reliance (percentage contribution to total household income) of livestock, crop and wage and business income of regular households at increasing distance from the PA boundary...... 75

xiii LIST OF PLATES

Plate 3. 1: Group discussion for participatory wealth ranking exercise in Robanda village ...... 37 Plate 3. 2: The researcher discussing with schoolchildren in Robanda primary school ...... 40 Plate 3. 3: The researcher interviewing a respondent in Kowak village ...... 41 Plate 4. 1: Drying bushmeat packages outside hunter‟s household in Robanda village...... 68 Plate 4. 2: Bushmeat package exposed for sun drying outside a regular household in Robanda village...... 68 Plate 4. 3: Bushmeat package (Impala) measured from hunter‟s household in Robanda village...... 69 Plate 4. 4: Bushmeat package (Thomson‟s gazelle) sold in one of the regular household at Robanda village...... 69

xiv LIST OF APPENDICES

Appendix 1: Questionnaire surveys for household meat consumption in villages adjacent the Serengeti National Park...... 107 Appendix 2: Questionnaire surveys for hunters and bushmeat traders in villages adjacent the Serengeti National Park...... 114 Appendix 3: Schoolchildren Interview Guide and Data record sheet for dietary recording ...... 122 Appendix 4: Data sheet for Bushmeat Packages ...... 123 Appendix 5: Data sheet for prices of animal meat foods ...... 124 Appendix 6: Ethical clearance certificate for conducting research in western Serengeti ...... 125 Appendix 7: Average market prices of different products in the study villages ..... 126 Appendix 8: Publications ...... 127

xv LIST OF ABBREVIATIONS AND ACRONYMS

CBD Convention on Biological Diversity CITES Convention on International Trade in Endangered Species of Wild fauna and flora CMS Convention on Migratory Species COSTECH Tanzania Commission for Science and Technology FAO Food and Agriculture Organisation of the United Nations FGD Focused Group Discussion GCA Game Controlled Area GLM Generalised Linear Model GPS Global Positioning System GSE Greater Serengeti Ecosystem HH Household NGO Non-Government Organisation NIMR National Institute for Medical Research NTNU Norwegian University of Science and Technology PA Protected Area SNP Serengeti National Park TAWIRI Tanzania Wildlife Research Institute TZS Tanzanian Shillings UDOM University of Dodoma URT United Republic of Tanzania VEO Village Executive Officer VIF Variance Inflation Factor WMA Wildlife Management Area

xvi CHAPTER ONE

INTRODUCTION

1.1 Background of the study Bushmeat is defined as meat from the wild animals hunted and processed for human consumption (Adefalu et al., 2012; Gideon, 2014; Kümpel, Milner-Gulland, Cowlishaw, & Rowcliffe, 2010; van Vliet, Nabesse, & Nasi, 2015). Bushmeat species include mostly vertebrate animals such as ungulates, primates, large-bodied birds and small mammals (Mwakatobe, Røskaft, & Nyahongo, 2012). Bushmeat hunting refers to harvesting of wild animals for food and non-food purposes (van Vliet et al., 2015) and it occurs mostly in areas where hunting is prohibited or without permits (FAO, 2015; Fischer, Naiman, Lowassa, Randall, & Rentsch, 2014). It also uses prohibited hunting methods and it kills protected species regardless of age and sex of the animal (Loibooki, Hofer, Campbell, & East, 2002). Bushmeat hunting is among the challenges facing conservation efforts as it causes depletion of wildlife populations in many locations in Sub-Saharan Africa (Milner-Gulland, Bennett, & Group, 2003; Ripple et al., 2016; Wilkie, Bennett, Peres, & Cunningham, 2011). Historically, humans have sustainably exploited wild animals for food by using traditional hunting tools (Costello, Burger, Galvin, Hilborn, & Polasky, 2008; Galvin, Polasky, Costello, & Loibook, 2008; Loibooki et al., 2002), because the tools were too crude to exploit the animals to unsustainable levels also the human population was relatively low (Ofori & Attuquayefio, 2010; Primack, 2010). At that time, bushmeat hunting was not considered a problem until the ingress of conservation movements during the 19th century (Primack, 2010). The human population growth as well as the use of advanced hunting tools such as wire snares and guns have contributed to unsustainable bushmeat hunting through overexploitation (Mendelson, Cowlishaw, & Rowcliffe, 2003; Milner-Gulland et al., 2003; Ofori & Attuquayefio, 2010; Primack, 2010; Schenck et al., 2006).

Bushmeat hunting and consumption are common among developing countries in Sub-Saharan Africa, Asia and Latin America influenced by cultural, socioeconomical and geographical factors (Chaves, Wilkie, Monroe, & Sieving, 2017; Nielsen,

1 Pouliot, Meilby, Smith-Hall, & Angelsen, 2017). The main reasons for hunting wild animals include acquiring meat for human consumption and other commercial purposes (Ripple et al., 2016).

In Sub-Saharan Africa, bushmeat constitutes an important source of household food and income, although the magnitude of exploitation varies across regions influenced by the availability of wild animals (Ahmadi et al., 2018; Barnett, 1997; Nielsen et al., 2017). It is particularly important to poor rural households (Coad et al., 2010; Fischer et al., 2014) providing protein and essential micronutrients that may otherwise be inaccessible with potentially severe health implications (Fa et al., 2015; Golden, Fernald, Brashares, Rasolofoniaina, & Kremen, 2011).

Many wildlife populations have already declined because of hunting and bushmeat trade, and the situation is worse in West and Central Africa (Emmanuel de Merode & Cowlishaw, 2006; Rentsch & Damon, 2013; Schulte-Herbrüggen, Cowlishaw, Homewood, & Rowcliffe, 2013; Wilfred, 2015). Bushmeat consumption is common both in rural and urban areas and it serves as an important source of meat protein in households (Kümpel et al., 2010; Mendelson et al., 2003; Schenck et al., 2006; van Vliet et al., 2015). It is also an important source of household income through trade particularly for poor households (Adefalu et al., 2012; Cowlishaw, Mendelson, & Rowcliffe 2005; Damania, Milner-Gulland, & Crookes, 2005; Kümpel et al., 2010). Bushmeat is sold together with other commodities in open markets (Cowlishaw et al., 2005; McNamara et al., 2016; van Vliet et al., 2017) and the income earned can easily be quantified (Kümpel et al., 2010).

The bushmeat trade is economically defined as the exchange of meat from hunted wild animals for financial or material gain (Rogan, Lindsey, & McNutt, 2015). It is among the common activities that are crucial for peoples‟ livelihoods (Crookes & Milner-Gulland, 2006; Edderai & Dame, 2006; Mendelson et al., 2003). The bushmeat trade operates through a well-organised network with various stakeholders such as hunters, wholesalers, retailers, buyers and middlemen (Mendelson et al., 2003). However, in East Africa, particularly Tanzania, the organisation of the bushmeat trade is not well known (Nielsen, Meilby, & Smith-hall, 2014), as it

2 operates in a black market because it is an illegal business. This makes difficulties in quantifying the contribution of bushmeat income to total household income.

Bushmeat hunting is still practised by communities adjacent to protected areas and other wildlife areas in Tanzania (Kideghesho, 2009; Nielsen, Jacobsen, & Thorsen, 2014; Wilfred, Milner-Gulland, & Travers, 2017). In western Serengeti, bushmeat hunting is an important source of income for primarily young men trading a third of their catch while the rest is consumed in the household (Loibooki et al., 2002). Estimates of the number of households engaged in bushmeat hunting vary between 8-57% of the total households (Nuno, Bunnefeld, Naiman, & Milner-Gulland, 2013). Most hunters come from nearby villages between 0 and 16 km from protected area boundaries, but some live as far away as 45 km (Kideghesho, 2010; Loibooki et al., 2002). Hunting occurs mostly during the dry season with high peaks in August to November, following the migration routes of animals, particularly wildebeest (Connochaetes taurinus) and zebra (Equus quagga) (Loibooki et al., 2002; Nyahongo, Holmern, Kaltenborn, & Røskaft, 2009).

About 66% of the people prefer bushmeat to other meat-protein food in their diet (Mwakatobe et al., 2012). Among others, the reason for this is the cultural aspect because bushmeat consumption are typically engrained in the customs and traditions of various tribes (Kideghesho, 2008). Depending on the nature of the rural economy, bushmeat is generally much cheaper than meat from domestic animals sold at prices between 0.85 and 1.0 US$ per kg (Rentsch & Damon, 2013). With such lower price together with cultural influences, the reliance for poor households on bushmeat consumption seems inevitable (Kiffner, Peters, Stroming, & Kioko, 2015; Ndibalema & Songorwa, 2007). Nevertheless, protein consumption is positively related to household wealth status (Rentsch & Damon, 2013).

The sustainability of extraction levels in the Greater Serengeti Ecosystem (GSE) is questionable, and hunting intensity is expected to increase further because of the rising human population in districts adjacent to the protected areas (Holmern, Muya, & Røskaft, 2007; Rogan et al., 2017; Setsaas, Holmern, Mwakalebe, Stokke, & Røskaft, 2007).

3 Therefore, there is an urgent need for understanding the role of bushmeat as food and income source for the households as well as the factors influencing household consumption and bushmeat income reliance in order to find alternatives. Such information can help in developing policy strategies that focus on controlling bushmeat hunting in order to ensure sustainable use of this wildlife resource (Luiselli et al., 2017). It is also important for predictions of the consequances associated with shortage of bushmeat to local people (Golden, Bonds, Brashares, Rasolofoniaina, & Kremen, 2014).

1.2 Statement of the Research Problem Previous studies have documented that bushmeat is an important food resource for rural communities in Sub-Saharan Africa (Barnett, 1997; Nielsen, Meilby, Smith- Hall, Pouliot, & Treue, 2018). Most of these studies have been conducted in West and Central Africa (Emmanuel de Merode, Homewood, & Cowlishaw, 2004; Fa et al., 2015; Milner-Gulland et al., 2003; Schulte-Herbrüggen et al., 2013). Despite being a widespread phenomenon, bushmeat hunting is poorly understood in the East African savannahs (Lindsey et al., 2013; Rogan et al., 2015) and has long been regarded primarily as a subsistence activity, and bushmeat trade thought negligible (CBD, 2011; Lindsey et al., 2013). Bushmeat hunting and consumption are also common in households located at various distances from protected area boundary on the western part of the Serengeti ecosystem in Tanzania (Kaltenborn, Nyahongo, & Tingstad, 2005; Loibooki et al., 2002; Nyahongo et al., 2009).

A considerable number of studies have attempted to quantify the importance of bushmeat to local communities (Fischer et al., 2014; Knapp, 2012; Mfunda & Røskaft, 2010) but still obtaining reliable data is complicated by the illegal nature of bushmeat hunting activities (Knapp, Rentsch, Schmitt, Lewis, & Polasky, 2010; Nuno et al., 2013).

Therefore, very little information exists on the importance of bushmeat to household meat consumption along a gradient of distance from PA boundary in western Serengeti and how it varies across seasons. Also the contribution of bushmeat income to total household income as a function of distance from PA boundary is still

4 unknown. Moreover, scarce information exists on the factors influencing bushmeat consumption in households and the factors influencing household participation in hunting and reliance on bushmeat income in western Serengeti.

Understanding the role of bushmeat and predictors of bushmeat hunting and consumption in households is crucial in developing appropriate conservation and development strategies for local people. Knowledge of the contribution of bushmeat to household meat consumption and income can help in the assessment of the alternative meat and income sources. If people depend more on bushmeat consumption, it is a threat to wildlife resources and also a challenge to conservation efforts. Moreover, if there will be no alternative sources of meat and income, people will continue to extract bushmeat unsustainably which in turn will affect the Serengeti ecosystem as well as the livelihoods of people that depend on it.

1.3 Objectives of the Study The general objective of the study was to assess the contribution of bushmeat to household food and income and factors influencing household dependence on bushmeat in western Serengeti.

1.3.1 Specific Objectives The specific objectives of the study were to: i. Determine seasonal variation in frequency of bushmeat consumption by households. ii. Quantify the contribution of bushmeat to household meat consumption as a function of distance from the SNP boundary. iii. Determine variations in the frequency of bushmeat consumption reported by schoolchildren and adults. iv. Determine the factors influencing bushmeat consumption in households. v. Quantify the contribution of bushmeat to household income and the influence of distance from PA boundary on bushmeat hunting activities. vi. Determine the factors influencing household participation in hunting and trading bushmeat and household bushmeat income reliance.

5 1.4 Hypotheses The study aimed at testing the hypotheses, that: i. Bushmeat consumption frequency is higher during the dry season than the wet season. ii. Bushmeat consumption frequency decreases with increasing distance from the protected area boundary. iii. Schoolchildren report higher bushmeat consumption frequency than adults. iv. Bushmeat consumption frequency is associated with household socioeconomic factors and the frequency of consumption of other animal meat sources. v. Bushmeat hunting and household reliance on bushmeat income decreases with distance from PA boundary. vi. Household participation in hunting and reliance on bushmeat income is influenced by household socioeconomic factors.

1.5 Significance of the Study The results from this study have shown the contribution of bushmeat to household income and food source in western Serengeti and how the consumption varies across seasons of the year. The study also identified factors influencing bushmeat consumption in households and participation in hunting and reliance on bushmeat income. All these information can help in the assessment of the alternative meat and income sources for the local people in such areas as it shows the extent to which people depend on bushmeat. This can help in formulating strategies to reduce the problem of illegal hunting which affects conservation efforts. Furthermore, the study has contributed on the methods and approaches to be used when conducting studies related to illegal hunting where respondents are hardly to find.

Moreover, information obtained from this study can serve as baseline information on the importance of bushmeat to households living adjacent protected areas particularly in western Serengeti ecosystem. Furthermore, the information obtained can help to inform the management authorities including policy-makers on the ways to find alternative sources of meat and income for households in western Serengeti. Results from this study can also help in the efforts to reduce the problem of illegal hunting

6 by concentrating on identified targeted areas, the communities close to protected areas. The alternative meat sources used by households‟ located far distances from the protected areas can serve as a lesson in the formulation of strategies to reduce bushmeat dependence in villages located close to protected areas.

Furthermore, modelling of the factors influencing household reliance on bushmeat has helped to show the significant predictors. The identified predictors of household reliance on bushmeat can serve as baseline information in finding alternatives. Also the information obtained has helped to recognise the accountability of the local people living close to protected areas in the management of natural resources. Moreover, the findings of this study can help to inform the government and development partners on the need for creating meaningful pattern of controlling bushmeat trade and conserving natural resources.

1.6 Organization of the Dessertation This dessertation is organised into six main chapters. Chapter one accounts for the general introduction of the study. It provides information on the background of the study, research objectives, research hypotheses and significance of the study.

Chapter two summarises the theoretical and empirical literatures related to the study. It also explains what is known about bushmeat hunting activities in western Serengeti and therefore identifies the research gap from the unknown. In this chapter, the conceptual framework describing the relationships between dependent, independent and intermediate variables are also explained.

Chapter three describes the approaches and methods used by this study in data collection and how such data were analysed. It gives a brief background of the study area and the selection criteria. The characteristics of the study villages are also highlighted in this chapter. It describes the research design employed by this study and the ethical procedures followed as well as the limitations and delimitations of the study.

7 Chapter four presents the results obtained from this study. The results are organised based on the study objectives. Figures, tables and plates were also used to summarise the information presented in this chapter.

Chapter five discusses the study findings and give interpretations. The discussion is organised based on the research hypotheses. It discusses and interprets information obtained by this study and also integrates the discussion with other similar studies else where.

Chapter six provides the conclusion and recommendations from this study and also highlights areas for further research.

8 CHAPTER TWO

LITERATURE REVIEW

2.1 Overview This chapter presents the theoretical and empirical reviews on the important information relevant to this study. It also presents the knowledge gap from the summarised information on what is known and identifies what is unknown about bushmeat hunting activities in western Serengeti. Finally, it explains the conceptual framework guiding this study describing the relationships between dependent, intermediate and independent variables.

2.2 Definition of Key Terms 2.2.1 Bushmeat Hunting Bushmeat hunting is an illegal exploitation of wild animals for human consumption (FAO, 2015; Fischer et al., 2014; Kümpel et al., 2010). Bushmeat hunting is done mostly for human consumption as a source of meat protein food (Adefalu et al., 2012; Loibooki et al., 2002) although to some people it is a source of income through bushmeat trade. Despite the importance of it, bushmeat hunting is regarded as unsustainable activity which can cause overexploitation of wildlife resources (Ripple et al., 2016).

2.2.2 Protected Areas Protected Area (PA) refers to a clearly defined area for conservation purposes, managed through legal or other effective means, to achieve the objectives (URT, 2007). PAs are intended to maintain functioning ecosystems which helps to maintain ecological processes, and provide socioeconomic benefits to local communities around, through sustainable utilisation of the resources protected (Dudley, 2008; Golden, Rabehatonina, Rakotosoa, & Moore, 2014; Masuruli, 2014).

2.2.3 Ecosystem Goods and Services Ecosystem goods and services are the benefits that humans obtain from the natural environment and mostly from a well functioning ecosystem which contribute directly or indirectly to human well-being (Primack, 2010). The ecosystem services are generally grouped into four major functions which are provisioning, regulating,

9 cultural, and supporting services. Bushmeat is one of the ecosystem goods provided by the ecosystem in terms of provisioning services.

2.2.4 Local Communities In this study, local communities are defined as residents living adjacent to PAs in the Serengeti ecosystem. The ecosystem is surrounded by many local communities with different ethinicity and background, socio-cultural, economic, environmental dimensions, levels of development, opportunities, and constraints (Masuruli, 2014).

2.3 Theoretical Review 2.3.1 The Social Exchange Theory Social exchange theory explains how peoples‟ decisions are influenced by cost- benefit analysis (Homans, 1958) as they intend to maximise benefits while reducing costs (Shogren et al., 1999). It involves social behavious which have economic relationships in a way that goods possessed by each party are economically valued by the other party. This theory implies that, people make decision based on cost-benefit analysis which is also applied to local people who decide to engage with bushmeat hunting activities in order to maximise individual benefits despite the cost of being arrested by park rangers (Nyahongo, East, Mturi, & Hofer, 2006). The hunting is illegal because wildlife resources are found inside protected areas and the benefits obtained from conservation are rarely observed as individual benefits but rather communal benefits (Nyahongo et al., 2006). However, most people do value individual benefits than communal benefits (Wilfred et al., 2017). Still communal benefits are more important as they benefiting the whole community and in most cases for a long period of time while the individual benefits foristance illegal hunting provide a short term benefit to an individual. Some of the communal benefits includes sharing arrangements and outreach programs providing social services including construction of schools, health centers and government offices, water supply, provision of health services and health insuarance, creation of employment opportunities, and also education sponsorship for students from poor families (Masuruli, 2014; Wilfred, 2010). However, other theories should also be used due to some limitations of this theory as it based on cost-benefit analysis of an individual

10 while there are also other factors that can motivate people to engage in bushmeat hunting.

2.3.2 The Rational Choice Theory The rational choice theory was developed based on the three main assumptions of neoclassical economics stated that; individuals have selfish preferences, they maximise their own utility and they act independently based on full information (Wittek, 2013). The theory explains the behaviour of people in relation to their preferences. This can help in understanding the society through individual behaviours as explained through rationality in which choices are consistent as they are made according to preferences (Ostrom, 1998).

Preferences can be defined as positive or negative evaluations that people attach to an outcomes of their actions or behaviours (Wittek, 2013). In rational choice approach, selfishness drive people to maximise their benefits even without considering the rules (Lindenberg, 2001). Based on this theory, actions chosen by the local people can help to determine the success or failure of conservation. Decisions about whether to do illegal hunting, grow crops, harvest fuel wood, or raise livestock impact the ecosystem.

The theory explains how peoples‟ decission are made with intention to maximise their benefits by choosing the best alternatives despite all the uncertainties (Mas- Colell, Whinston, & Green, 1995; Ostrom, 1998). Some of the decisions are beneficial for conservation for example, set aside community land for conservation purposes. However, some people make decision to engage in activities that degrade the environment (Acheson, 2002). For example, an individual can choose to hunt bushmeat illegally, which yields immediate benefits to the individual, but can cause population decline and sometimes extinction. On the other hand, an individual can choose not to hunt bushmeat, forgoing immediate consumption benefits in order to have long term benefits.

2.4 Empirical Review Empirical literature review summarises the information related to this study based on experiences and observations, rather than systematic logic or theories. This section

11 highlights key information about bushmeat hunting activities observed and measured phenomena and therefore, develops knowledge from real experience rather than from theories.

2.4.1 Tanzania Wildlife Resources Management Policy Frameworks Wildlife management involves policies and legislation that govern the utilisation of wildlife resources in terms of hunting and trapping of animals as well as trade of wild animals or products. Policies and laws helps in the protection of wildlife habitats and restoration of degraded ecosystems as well as protection of species that are threatened or in danger of extinction (URT, 2007). Wildlife protection policies and legislations span a wide range from the national level to the international level. The international level reflects the multinational movement of some migratory species and the increasing recognition that environmental conservation, including protection of wildlife, is a global responsibility (CBD, 2011).

The government of Tanzania has formulated different policies that work together to safeguard the wildlife resources. The main aim is to ensure sustainability and to enhance protection and conservation of wildlife resources and their environments. Wildlife is part of the natural resources found in Tanzania and its protection and conservation is influenced by wildlife policies as well as other policies related to management of the environment and natural resources. The Tanzania wildlife resources management policy frameworks includes; The National Environmental Policy (1997), The National Wildlife Policy (2007), The National Forest Policy (1998) The National Tourism Policy (1999) and The National Land Policy (1997). In order to enforce these policies, the government has national laws such as; The Environmental act (2004), The Wildlife act (2009), The Forest act (2002), The National park act (2009), The Tourism act (2008) and The Land act (2004).

2.4.1.1 The National Environmental Policy The environmental policy of Tanzania intends to consider the environment in any decision making and to provide for sectoral and cross-sectoral policy analysis in order to achieve compatibility among sectors (URT, 1997a). There are other policies that can affect wildlife in their actions, therefore the environmental policy act as the

12 main policy guiding other policies in the management of the environmental resources. The policy objectives strengthen on the conservation of biodiversity and sustainable use of natural resources, environmental protection and pollution prevention and control (URT, 1997a). The policy also intends to increase awareness and community participation in environmental actions and this is very important in achieving conservation objectives. Sustainable use of wildlife is among the objectives of conservation and the local communities should support this in order to reduce illegal bushmeat hunting.

However, the environmental policy of Tanzania is still facing challenges in addressing inter-sectorial issues which affects wildlife for example; issues related to land uses cut across different policies. The implementation of such policy objectives can affect wildlife in one way or another although most of these activities are important for peoples‟ livelihoods and development of the country. For example road construction across national parks and other protected areas is a threat to wild animals due to road accident and edge effects associated with habitat fragmentation. This can also increase the probability of illegal bushmeat hunting activities in the area (Walelign, Nielsen, & Jacobsen, 2019).

2.4.1.2 The National Wildlife Policy The policy highlights the importance of wildlife resources including biological values of species and habitats such as the Serengeti ecosystem and others that are found in Tanzania (URT, 2007). It also account for the economic value of wildlife resources and their potential to contribute to the sustainable development of Tanzania. Potential wildlife areas are legally protected and categorised into different protection and management status. The national wildlife policy of Tanzania has four different categories of protected areas such as National parks, Game reserves, Game Controlled Areas (GCAs) and Wildlife Managed Areas (WMAs). In National parks and Game reserves no human settlement is allowed while in GCAs and WMAs human and wildlife co-exist. In such areas where human interact with wildlife, there is a big challenge of human wildlife conflict because of crop raiding, livestock depradation and human attack by wild animals (Holmern, Nyahongo, & Røskaft, 2007). On the other hand people also do illegal exploitation of natural resources and

13 retaliating killing of wild carnivores. However, illegal bushmeat hunting is also practiced in national parks and game reserves and the Serengeti National park is among the victims (Loibooki et al., 2002; Nyahongo et al., 2006; Rentsch, Hilborn, Knapp, Metzger, & Loibooki, 2015). Despite all these challenges, the policy objective is to ensure equitable sharing of benefits arising from wildlife utilisation and provide compensation for wildlife related costs. However, wildlife sector is still facing some challenges that have contributed in failure of eliminating illegal bushmeat hunting and other illegal activities as follows;  Failure of wildlife conservation as a form of land use to compete adequately with other land use, especially to the village communities.  Persistence illegal taking of wildlife, wildlife trade and increased problem animals.  Inadequate wildlife use right especially to the village communities.  Low budgetary allocation for conservation and development of the wildlife sector at the local government level.  Poor infrastructure in wildlife areas and limited human resource to carry out wildlife conservation activities.

2.4.1.3 The National Forest Policy The general aim of the forest policy is to enhance the contribution of forest sector to the sustainable development of the country and the conservation and managemet of natural resources for the benefits of the present and future generations (URT, 1998). The policy promotes conservation of ecosystems and biological diversity appropriate management and utilisation methods. About a quater of Tanzania land is covered by unique ecosystems in the form of forest reserves, national parks and game reserves. These forms of protected areas help to conserve wildlife and the environment in general and therefore enhance sustainable utilisation of the resources available. In such areas there is abundance of wildlife including big game which supports tourism industry as a source tourist attraction. The policy intends to incorporate wildlife in the forest management plans and set aside areas for wildlife corridors so as to ensure habitats for different types of wildlife. The policy also promotes involvement of local communities and other stakeholders in conservation and management of forest

14 resources and other natural resources. The involvement of local people is very important in order to increase awearness since some of their activities are not environmentally friendly and can deteriorate wildlife popupations (Nindi, 2010). Such activities include illegal logging, enchroachment, wildfires, overgrazing and illegal bushmeat hunting.

However, among the challenges facing forest sector in Tanzania is inadequate mechanism for enhancing inter-sectorial coordination (URT, 1998). Wildlife management was not incorporated in the current forest managemet plan and some forest reserves overlap with game reserves or game controlled areas causing conflicts in management activities. This is a challenge to wildlife managers especially in controlling illegal bushmeat hunting because poachers can use permits from forest management to get access to wildlife areas.

2.4.1.4 The National Tourism Policy Tanzania tourism potentials include wildlife resources and other attractions such as landscape, scenery, water bodies and beaches, a diversity of cultures and numerous archeological sites (URT, 1999, 2007). More than 44% of the land is set aside for protected areas in Tanzania and among the biggest tourist attraction in Africa; three are found in Tanzania including the Serengeti National Park, Ngorongoro Conservation Area and Mount Kilimanjaro. Attraction from wildlife is among the leading contributors of benefits accrued from tourism industry in Tanzania (URT, 1999). The policy recognises the role of conservation institutions which are to enhance conservation and awareness and to ensure proper management of the natural resources. This can help to control illegal bushmeat hunting and other illegal activities and therefore promote sustainable use of wildlife resources. The policy promotes local people involvement in the development and management of tourist projects in order to increase benefit sharing of income generated from tourism activities. It gives priority to community members in terms of training and employment and other socioeconomic benefits occurring from tourism investiments (URT, 1999). All these help to increase conservation benefits to people as a way to reduce poaching and get support for conservation efforts.

15 Despite the benefits highlighted in the policy, some people still have negative attitudes towards conservation and therefore engage in illegal bushmeat hunting in order to get more individual benefits. Illegal bushmeat hunting can cause population declines and therefore affects tourism industry in Tanzania which inturn can affect the economy of the country. Tourism policy needs to develop strategies in collaboration with other policies to control illegal bushmeat hunting.

2.4.1.5 The National Land Policy The main policy objective is to enhance a secure land tenure system and sustainable use of the land resources (URT, 1997b). The long-term environmental protection can be achieved when people have secure land ownership, access to environmental resources and right to use them. The policy intends to collaborate with other sectors in relation to management of land resources in order to use institutional framework in solving conflicts on the use of land resources (URT, 1997b).

The land use conflicts is a growing challenge in Tanzania due to disorganised resource allocation as well as rapid human population growth (Nindi, 2010). Overlapping land uses in game controlled areas allows other activities such as agriculture and settlement to occur simulitaneously which results into conflicts. The game controlled areas are critical habitats for wildlife species at a particular season or corridor for migratory animals. Potential and sensitive areas such as wildlife corridors or migratory routes and other critical areas as habitat for certain species should be highly considered for protection since illegal bushmeat hunters also target these areas.

2.4.2 Impelementation of International Wildlife Policy Framewok in Tanzania The government of Tanzania has also ratified some of the international treaties and conventions in order to implement the international policies and laws governing the management of wildlife resources. The international policies related to wildlife resources management in which Tanzania is implementing includes; Convention on Biological Diversity (CBD) (1992), Convention on International Trade on Endangered Species of Wild Fauna and Flora (CITES) (1975), Convention on migratory species (CMS) and The Ramsar Convention (1975).

16 2.4.2.1 The Convention on Biological Diversity The Convention on Biological Diversity (CBD) is an international agreement which aim to conserve biodiversity, ensure sustainable use of biodiversity and enhance fair and equitable sharing of the benefits arising from utilization of genetic resources. The CBD assists its members in identifying and implementing strategies to conserve biodiversity through scientific research, economic incentives and legal means. It has helped to increase awareness of the importance of wildlife protection at the global level. The policies and laws concerned with wildlife protections that are in effect globally helps to conserve wildlife and their habitats and also threatened species. Tanzania is a member of CBD since 1997 (URT, 2007) and it has formulated policies that implement CBD objectives in order to conserve biodiversity and to ensure sustainable use. The SNP is one of the protected areas found in Tanzania and this is a form of in-situ conservation in which the CBD intends to achieve. Through protected areas wild animals are protected against illegal hunting although efforts to control poaching is expensive in developing countries and therefore requires support from developed countries particularly member states.

2.4.2.2 Convention on International Trade in Endangered Species of Wild Fauna and Flora The Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) is a multilateral agreement between states with the aim of ensuring that trade in wildlife both plants and animals is controlled internationally in order to promote sustainability of the species involved. Tanzania became a member of CITES in 1981 (URT, 2007), since that time it works to implement the objectives through national policies and regulations governing the management of wildlife trade. It has helped to control illegal trophy hunting which highly affected populations of elephants and rhinos in 1980s for commercial purposes (Kideghesho, 2010).

The level of illegal trophy hunting has reduced in Tanzania and the remained problem is illegal bushmeat hunting. Illegal bushmeat hunting and consumption is increasing through bushmeat trade which is common in Sub-Sahara Africa (Lindsey et al., 2013; Nielsen et al., 2018). Bushmeat trade is among the challenge that requires international attention since some countries have remained with few wild

17 animals that can be used as source of bushmeat and therefore will depend on other countries with higher abundance.

2.4.3 Wildlife Resources Utilization in Tanzania The national wildlife policy identified different forms of wildlife utilization such as game viewing, tourist hunting, resident hunting, wildlife farming, breeding and ranching, eco-tourism, Zoos and game sanctuary (URT, 2007). Game viewing which is currently practiced in the Serengeti National Park and in other national parks has a great economic potential for earning both local and foreign currency and also a source of employment. Game viewing or photographic tourism is the most important form of non-consumptive wildlife utilisation in Tanzania. The consumptive wildlife utilization requires lincense and principally there are two forms of hunting; tourist hunting and resident hunting (URT, 2007). The resident hunting is restricted to Tanzanian citizens and resident foreigners and it should follow the regulations as specified in resident hunting regulations of 2012. It requires hunting licence and the hunting should be done using specified hunting tools including guns which are rarely possessed by local Tanzanians.

Despite the legal opportunity for resident hunting, most local people do illegal hunting because majority of them fail to meet the required conditions to get hunting permit or licence. Illegal bushmeat hunting also occur widely, although trophy hunting is often said to be less practiced than it was in the 1970s and 1980s before the nation wide ant-poaching campaign (Kideghesho, 2010; Walsh, 2006). Sustainable utilisation of wildlife resources can substantially contribute to an increased in export and domestic economy through tourism, diversification of economic activities, sustaining food security and direct increase production in other sectors such as fishery, agriculture, transport and support to the informal sectors (URT, 2007).

2.4.3.1 Status and extent of Bushmeat Hunting in Tanzania Over 70% of the Tanzanians live in rural areas and most of the communities depend on environmental resources for their livelihoods (URT, 2007). In such areas wildlife also exist and about 6% of the total area is occupied by wildlife. This means that

18 people and wildlife do interact in some areas and this pose challenges to both people and wildlife. The greatest threat to wildlife populations is the illegal hunting done by local people while crop raiding and livestock depredation are the costs incurred by people caused by wild animals. Bushmeat hunting and consumption is common in Tanzania particularly in areas adjacent protected areas (Andimile, Tim, & Mulder, 2012; Knapp, Peace, & Bechtel, 2017; Manyama, Nyahongo, Nielsen, & Røskaft, 2019; Nielsen & Meilby, 2015). The wildlife authorities have the mandate to combat illegal taking of wildlife resources by a dual but interlinked strategy which focuses both inside and outside wildlife protected areas (URT, 2007). However, this is a difficult task that requires resources in terms of adequate staff, equipements and fund. It also requires great intelligence since poachers develops new strategies to run away from management authorities.

2.4.4 The Importance of Bushmeat Bushmeat has significant role in households as a source of food and income particularly in Sub-Saharan Africa (FAO, 2014). This shows how wildlife is important in peoples‟ livelihoods since it is the main source of bushmeat and some wildlife products are used for medicinal and spiritual purposes across societies (Morsello et al., 2015; Primack, 2010). Besides, bushmeat hunting and consumption is also important for some people as it portray their culture and therefore give identity of indigenous people (Kideghesho, 2008). Bushmeat hunting is crucial activity for households as it serves as a source of meat protein in the diet of many people in rural areas (Brashares, Golden, Weinbaum, Barrett, & Okello, 2011; Schenck et al., 2006). Bushmeat substitutes other meat protein sources which are considered expensive and it provide nutrient and therefore, enhances food security (Fa, Currie, & Meeuwig, 2003; Nyahongo et al., 2009).

Bushmeat is also important in household economy as a source of income through the bushmeat trade (Damania et al., 2005; Jacob, Nelson, & Ukpong, 2018; Manyama, Nielsen, Røskaft, & Nyahongo, 2019; Nielsen et al., 2017). It is the livelihood opportunity which is used mostly during times of hardship when other alternative income sources have failed (Fa & Brown, 2009; Nielsen & Meilby, 2015). However, livelihood studies conducted in Sub-Saharan countries shows that income

19 diversification is widespread, and the importance of farm and off-farm income, of which bushmeat is a part, varies greatly across localities (Knapp, 2007; Kümpel et al., 2010).

Despite being an important resource to most people, bushmeat hunting is a threat to wildlife due to overexploitation driven by high consumption levels. Unsustainable bushmeat hunting is among the major causes of species decline across the world (Lindsey et al., 2013; Ripple et al., 2016). In most regions where hunting has been studied, vertebrates contribute almost all of the wild meat consumed and traded. Recent estimations suggest that nearly six million tons of meat from wild mammals may be eaten in the Afro-tropics each year (Nasi, Taber, & van Vliet, 2011). Consequently, about 301 species of terrestrial mammals are threatened with extinction from hunting for food and medicinal products (Ripple et al., 2016).

2.4.5 Bushmeat Hunting in Western Serengeti The SNP is a large area with high diversity of species as well as habitats, and it is characterised by the annual migration of wildebeest and a large diverse community of herbivores (Kideghesho, 2010; Manyama, Nielsen, et al., 2019). The animals migrate without boundaries and therefore, exposed into risks of being hunted when they reach open areas where local people lives (Sinclair, Metzger, Mduma, & Fryxell, 2015). For instance, migratory herbivores particularly wildebeest, are mostly hunted ranging from 97,796 - 140,615 individuals per year (Rentsch & Packer, 2015), due to their high abundance and availability during the migration.

Bushmeat species range from insects, reptiles and birds, to large mammals depending on their availability in a particular area. The type and size of the animal consumed suggest the conservation status and species abundance in the area. In places where insect and birds are highly consumed (Manyama, Nyahongo, & Røskaft, 2014), the populations of large mammals have already declined. In areas where most of the wildlife are found in protected areas, illegal bushmeat hunting is among the ways used by local communities to obtain benefits from wildlife conservation (Ceppi & Nielsen, 2014; Kideghesho, 2010).

20 The local people in western Serengeti consider bushmeat hunting as an important source of both food and income (Loibooki et al., 2002; Rentsch et al., 2015), and they consider it healthier and a delicacy than domestic meat (Ceppi & Nielsen, 2014). However, income generation from bushmeat trade is proved to be the main driver for illegal bushmeat hunting which is mostly done by men in the western Serengeti (Loibooki et al., 2002). Most of the hunting occurs in the dry season when the large herds of migratory herbivores particularlly wildebeests and zebras arrive from their migration (Manyama, Nielsen, et al., 2019).

Illegal bushmeat hunting occurred in the Serengeti ecosystem since 1960s mainly for food at household levels (Arcese, Hando, & Campbell, 1995). The commercial hunting for trophy was unnoticed until the 1970s and 1980s following the increased incidents of poaching that led to declines in populations of charismatic and keystone species, such as elephant (Loxodonta africana), black rhinoceros (Diceros bicornis), and buffalo (Syncerus caffer) (Arcese et al., 1995; Campbell & Hofer, 1995; Kideghesho, 2010; Sinclair, Packer, Mduma, & Fryxell, 2008). Although illegal hunting for trophies has already declined to almost a level of elimination, that of bushmeat has remained a major challenge because it is still practised in the area (Fischer et al., 2014; Sinclair et al., 2015). The local people kill large numbers (70%) of migratory animals when they move en route to northern Serengeti (Rentsch et al., 2015). The meat is sundried and sold or exchanged for other household commodities brought by vendors from distant villages (Mwakatobe et al., 2012).

2.4.6 The Bushmeat Market Structure and Trading System Bushmeat hunting is done for mainly two purposes which are substance use or household consumption and commercial use in order to earn household income (Mfunda & Røskaft, 2010; Nyaki, Gray, Lepczyk, Skibins, & Dennis, 2013). Subsistence bushmeat hunting is mainly for household food and it is common in many rural communities (Fischer et al., 2014). It uses traditional hunting methods, such as spear, domestic dogs, wire snares, pitfall traps, fire, poisoning, dazzling by torchlight, and bow and arrows (Mfunda & Røskaft, 2010). However, some of the hunting methods such as wire snares and pitfall traps are discouraged by conservationists because they kill non-targeted animals without considering the

21 number, age and sex (Gideon, 2014; Loibooki et al., 2002). Such methods are also used in western Serengeti particularly wire snares (Knapp, 2012; Manyama, Nielsen, et al., 2019; Nyahongo et al., 2006; Rentsch et al., 2015).

Hunting for commercial purposes is done through organised market structure and uses more hunting gears including firearms (Mfunda & Røskaft, 2010; Mwakatobe et al., 2012). It involves a chain of stakeholders such as hunters, wholesalers, retailers and consumers. Commercial hunting is regarded as an economic activity (Ndibalema & Songorwa, 2007; Rentsch & Damon, 2013) since it contributes as a source of income for the household through bushmeat trade (Knapp, 2007; Rentsch et al., 2015). The bushmeat trade has expanded due to increased market demand both in villages and towns in Sub-Saharan Africa (FAO, 2015; McNamara et al., 2016; Walelign, Nielsen, & Jakobsen, 2019). Bushmeat has a good market both in rural and urban communities, which act as a market force driving poachers to hunt (Rentsch & Damon, 2013). In order to control the illegal bushmeat trade, it is important to understand the bushmeat market structure, the commodity chains and finally the uses both locally and internationally (Shively, Jagger, Sserunkuuma, Arinaitwe, & Chibwana, 2010). Several approaches can be used to study the bushmeat trading system depending on the nature of the study site.

Participatory observations, interviews and household questionnaire surveys through snowballing techniques can be used if the business is illegal (Knapp, 2012; Nielsen, Meilby, et al., 2014). A clear understanding of the bushmeat market structure from the bushmeat commodity chain analysis will help in formulating measures to control illegal bushmeat activities (Brooks, Kebede, Allison, & Reynolds, 2010; Nielsen, Meilby, et al., 2014). A commodity chain refers to a collection of activities such as production, processing, transportation or distribution, final sale and end use of a particular commodity (Coad et al., 2010; Ribot, 1998). Previous studies have identified some of the key actors in the bushmeat commodity chain such as hunters who hunt the animals and supply bushmeat to traders including wholesalers who purchase bushmeat from hunters and sell the meat in bulk to retailers, who sell bushmeat to consumers (Nielsen, Meilby, et al., 2014). A clear understanding of the

22 bushmeat commodity chain in the Serengeti can help to formulate strategies to control the illegal bushmeat trade.

2.4.7 Factors Influencing Bushmeat Hunting Generally illegal behaviours such as bushmeat hunting are influenced by several factors including cultural, socioeconomic and geographical factors. Bushmeat hunting in western Serengeti is also influenced by many factors such as cultural, socioeconimc and geographical factors (Fischer et al., 2014; Loibooki et al., 2002; Nyahongo et al., 2009). The geographical location of the Serengeti and the surrounding villages also influences people to depend on bushmeat hunting and consumption in the area. Although bushmeat hunting is done mainly for subsistence and commercial purposes, other factors have contributed to its popularity. Such factors include human population growth, poverty and inadequate alternative meat sources (Kideghesho, 2010; Rentsch & Damon, 2013).

Poverty acts as a driving force for poor people who mostly hunt because of food insecurity and also a need for cash income to sustain their livelihoods (Knapp, 2007; Manyama, Nielsen, et al., 2019; Ndibalema & Songorwa, 2007). A large proportion of people in western Serengeti are economically poor as they live under 1 US$ per day and have an average of 150 US$ annual income per household (Barrett & Arcese, 1998; Kideghesho, 2010). Previous studies have found that illegal bushmeat hunting is conducted mainly by poor people in the communities (Fischer et al., 2014; Loibooki et al., 2002; Ndibalema & Songorwa, 2007). However, the contribution of poverty in bushmeat hunting is debatable because some few rich people are also involved in illegal bushmeat activities.

The human population growth accelerates the demand for bushmeat both in villages as well as in big towns (Mfunda & Røskaft, 2010; Ndibalema & Songorwa, 2007; Nyaki et al., 2013; Rentsch & Damon, 2013; Walelign, Nielsen, & Jakobsen, 2019). The human population in communities around the Serengeti ecosystem is more than 2 million with an average growth rate of 3.5% annually (Estes, Kuemmerle, Kushnir, Radeloff, & Shugart, 2012; URT, 2013). Apart from natural increase, the human population grows through immigration due to good agricultural land for cultivation

23 and livestock keeping, wildlife for bushmeat and tourism activities, gold deposits and water bodies (e.g., Lake Victoria and Mara River) for fishing (Kideghesho, 2010). Increased human population, particularly in rural areas increases the dependence on bushmeat (Kideghesho, 2009) especially to large households with high numbers of household members from ≥ 5 living and eating together.

Inadequate alternative meat sources is also a factor that influences bushmeat hunting in many areas in Sub-Saharan Africa (Nielsen et al., 2018; van Vliet, Nabesse, Gambalemoke, Akaibe, & Nasi, 2012). The shortage of meat protein foods may be associated with the two factors already discussed in addition to other causes. The high dependence on bushmeat may result in unsustainable bushmeat hunting, which significantly affects wildlife populations, and in turn affects peoples‟ welfare. Moreover, inadequate law enforcement is a common problem in developing countries particularly in African and is also a driving factor for illegal bushmeat hunting (Holmern, Muya, et al., 2007; Mfunda & Røskaft, 2010; Nyaki et al., 2013).

2.4.8 Approaches in Studying Illegal Bushmeat Hunting Studying illegal bushmeat hunting activities is crucial for management purposes in order to find control measure to reduce or eliminate such activities. Illegal behaviours affects management decisions due to uncertainties associated with the illegal activities done by people secretly (Nuno et al., 2013). Bushmeat hunting activities are hardly to study because of the illegal nature of the activities which makes people fear of repraisal. A couple of methods can be used depending on the study objectives, location, capacity and budget (Gavin, Solomon, & Blank, 2010). Such methods include direct interviews, dietary recall surveys, market surveys, self reporting, arrest records from law enforcement and indirect interview approaches (Fischer et al., 2014; Knapp et al., 2010; Loibooki et al., 2002; Nuno et al., 2013). However, each method has its merits and demerits, for example arrest records from law enforcement only acquare information from arrested hunters and not hunters who were not arrested as well as non-hunters (Fischer et al., 2014). Moreover, dietary recall survey which is considered a reliable method (Brashares et al., 2011; Knapp et al., 2010) has a challenge in distinguishing between purchased bushmeat and bushmeat hunted by household members.

24 Direct questioning method is less expensive and can be used in many studies but in reality it is a challenge when dealing with illegal activities. Asking respondents directly about illegal behaviours is a difficult task which requires other strategies in order to built trust among respondents (Fischer et al., 2014). Despite that, respondents can still refuse to share the information and this may lead to a non- random selection of respondents (Nuno et al., 2013). Indirect methods can also be used in such situations through a randomised response and therefore increases the accuracy of the information collected as well as reducing bias (Warner, 1965).

2.5 Knowledge Gap Bushmeat hunting and consumption have been documented in several studies in Sub- Saharan Africa (Brashares et al., 2004; Ceppi & Nielsen, 2014; Loibooki et al., 2002; Nielsen et al., 2018; van Vliet & Mbazza, 2011). Such studies have been concentrated on assessing whether people are doing illegal hunting as well as the reasons for people to engage in such activities particularly in Tanzania (Andimile et al., 2012; Bitanyi, Nesje, Kusiluka, Chenyambuga, & Kaltenborn, 2012; Fischer et al., 2014). However, none of these studies have reported the amount of bushmeat consumed by households as well as the contribution of bushmeat income to total household income along a gradient of distance from protected area boundary. Furthermore, information on the factors influencing bushmeat consumption and household reliance on bushmeat income is also lacking.

Understanding the importance of bushmeat to households is essential in developing appropriate alternatives that will help to enhance conservation and development strategies. However, obtaining such information is complicated due to the nature of the activity being illegal (Fischer et al., 2014; Nuno et al., 2013) which means that respondents are reluctant to share information for fear of reprisals (Knapp et al., 2010). Many studies have used adult people to collect information about bushmeat hunting and consumption in western Serengeti (Kaltenborn et al., 2005; Loibooki et al., 2002; Mwakatobe et al., 2012; Nyahongo et al., 2009; Rentsch & Packer, 2015). However, some of adult respondents do hide the information (Fischer et al., 2014; Kiffner et al., 2015) even if they consume bushmeat daily or weekly.

25 In order to overcome the challenge of adults‟ fear in sharing sensitive information, some few studies have used children as respondent in gathering such information (Golden et al., 2011; Haule, Johnsen, & Maganga, 2002; van Vliet et al., 2015). Following appropriate ethical guidelines aiming to protect respondents, children may be a rich source of information about various questions concerned with household food and nutritional security (Baranowski et al., 2012; Maseko, Shackleton, Nagoli, & Pullanikkatil, 2017; McPherson, Hoelscher, Alexander, Scanlon, & Serdula, 2000). Therefore, this study intended to quantify the importance of bushmeat to household meat consumption by using both schoolchildren and adult respondents. Furthermore, the study intended to determine the contribution of bushmeat income to total household income and factors influencing bushmeat consumption and household reliance on bushmeat income in western Serengeti.

2.6 Conceptual Framework The relationships between dependent, independent and intermediate variables are described in the conceptual framework (Figure 2.1). The conceptual framework for this study is based on concepts from the empirical studies related to bushmeat hunting and consumption in western Serengeti (Kaltenborn et al., 2005; Knapp, 2007; Loibooki et al., 2002; Mwakatobe et al., 2012; Nyahongo et al., 2009; Rentsch & Packer, 2015). It explains the role of bushmeat in households and factors influencing household dependence on bushmeat income and consumption in western Serengeti. The independent variables or predictors of household dependence on bushmeat consumption and income includes distance to the SNP boundary (km), season of the year (dry or wet), household wealth rank (poor, middle or rich), status of the respondent (hunter, bushmeat trader or consumer), age, gender and the consumption of other meat sources (domestic meat and fish).

Household reliance on bushmeat income is more likely in households close to SNP and hunting frequencies vary with season following the migration of herbivores particularly wildebeest and zebra (Loibooki et al., 2002). It is also influenced by socioeconomic factors such as household wealth, age and gender, where females are less involved in hunting compared to males. Wealthier households are less dependence on bushmeat income as they rely on other income sources for their

26 livelihoods. Alternative meat source (fish and domestic meat) reduces household dependence on bushmeat consumption (Nyahongo et al., 2009). All these factors and their interactions influence household reliance on bushmeat consumption and income in western Serengeti.

Independent variables Intermediate variable Dependent variables

Distance from the Household village to SNP reliance on boundary bushmeat income Season (dry or wet) Participation (Income source) in hunting and bushmeat

Household wealth trade Household reliance on bushmeat consumption Age and gender of (Food source) household head

Consumption of other animal meat sources

Figure 2. 1: A Conceptual framework describing the role of bushmeat in households and factors influencing household consumption and bushmeat income reliance in western Serengeti.

27 CHAPTER THREE

METHODOLOGY

3.1 Overview This chapter describes different approaches and methods used in data collection and analysis. It provides information on the study area description and selection criteria, research design, data collection methods, data analysis, validity and reliability, ethical issues and limitations and delimitations of the study. This chapter describe the selection criteria for study participants and how they were sampled. It also explains the methods used to analyse the data and the statistical software package used.

3.2 Description of Study Area The study area was the Greater Serengeti Ecosystem (GSE) and the study villages (Robanda, Rwamkoma and Kowak) were selected in the western Serengeti (Figure 3.1). The GSE is a highland savannah region dominated by acacia woodlands in the west, riverine forests in the northern part and an altitude of 1,000–1,800 m above sea level (Mmassy, Fyumagwa, Bevanger, & Røskaft, 2018). It is a large ecosystem composed of various protected areas such as SNP (14,763 km2), which is located between 1o28‟– 3o17‟S and 33o50‟–35o20‟E in Tanzania. In addition to the SNP, GSE also includes the Ikona Wildlife Management Area (WMA) (600 km2), Ikorongo Game Reserve (563 km2), Grumeti Game Reserve (416 km2), Kijereshi and Maswa Game Reserves (2,200 km2) in the Southwest, and Ngorongoro conservation area (8,292 km2) and Loliondo game controlled area (4,000 km2) to the East.

The GSE is a World Heritage Site and ecologically significant area with higher abundance and diversity of animals which attract tourists from many parts of the world (Kaltenborn et al., 2005; Kideghesho, 2010). The GSE is well known world wide due to migration of herbivores (wildebeests and zebras) commonly known as “great wildebeest migration” which occurs every year. The GSE have large populations of resident wildebeest and other herbivores such as Giraffe (Giraffa camelopardalis), African buffalo (Syncerus caffer), Topi (Damaliscus korrigum), Elephants (Loxodonta africana), Impala (Aepyceros melampus), Eland (Tragelaphus oryx), Warthog (Phacochaerus aethiopicus), Grant‟s gazelle (Gazella grantii), Hippo

28 (Hippopotamus amphibius), and others (Kideghesho, 2010; Rentsch et al., 2015). The area also support large populations of carnivores such as Lion (Panthera leo), Spotted hyena (Crocuta crocuta), Leopard (Panthera pardus), Black-backed jackal (Canis mesomelas), and Cheetah (Acinonyx jubatus) as well as a high diversity of bird species (Kaltenborn, Nyahongo, & Kideghesho, 2011).

The western Serengeti is ecologically significant area because it serves as a buffer zone for SNP and also a corridor for migratory animals (Mwakatobe, 2013). The animals migrate through western corridor which also covers the village lands and therefore provide opportunity for bushmeat hunting which increases the availability of bushmeat for household consumption (Kaltenborn et al., 2005; Mwakatobe et al., 2012; Nyahongo et al., 2009). The study villages were selected purposely from three different districts (Serengeti, Butiama and Rorya) all from Mara Region, based on the established transect along a gradient of distance from SNP boundary on the western side (Figure 3.1). The first village namely Robanda, was about 3 km from the SNP boundary and was referred as the closest village followed by the intermediate village namely Rwamkoma, located at 27 km and the distant village namely Kowak located at 58 km. The districts have different biophysical, cultural and socio-economic characteristics and therefore serve as a representative sample of local communities living in the western Serengeti. The ethnic composition in the area is diverse including Ikoma, Kurya, Ikizu, Isenye, Zanaki, Jita and Luo tribes.

The area has a rapid human population growth at a rate of 3.5% annually (Estes et al., 2012; URT, 2013) and most people are economically poor live under 1 US$ per day (Kideghesho, 2010; Loibooki et al., 2002). The main economic activities are farming, pastoralism, agro-pastoralism, poultry, hunting, fishing, charcoal making and making local brews (Kideghesho, 2010; Loibooki et al., 2002; Manyama, Nielsen, et al., 2019). The main cultivated crops are maize (Zea mays), cassava (Manihot utilissima), finger millet (Eleusine coracana) and sorghum (Sorghum vulgare) and livestock kept are cattle (Bos taurus), goat (Capra hircus) and sheep (Ovis aries) (Kyando, Nyahongo, Røskaft, & Nielsen, 2019).

29

Figure 3. 1: Map of the study area and its location in Tanzania (left) showing the study villages Robanda, Rwamkoma and Kowak indicated with black dots.

3.3 Description of Study Villages The description of the study villages includes geographical location, climate, vegetation type, land use and demographic characteristics. Geoghraphically, the study villages were located in the western Serengeti at various distances from the SNP boundary. The climatic condition slightly differs between villages particularly on rainfall patterns and temperature. The average minimum temperatures ranges from 13°C to 19°C and maximum temperatures ranges between 25°C to 32°C annually (Campbell & Hofer, 1995). The rainfall pattern has two major seasons which are short rains that start in late November to January and the long rains from

30 March to May, with an average annual rainfall ranges from 600 to 1,200 mm (Mramba, Andreassen, & Skarpe, 2017; Norton-Griffiths, Herlocker, & Pennycuick, 1975). Generally, the area has two main seasons, dry and wet, although the exact onset of each season varies annually. The dry season, starts from June to October and the wet season from November to May (Mmassy et al., 2018; Norton-Griffiths et al., 1975). The dry season is further divided into short dry (January - February) and long dry (June - October) season, while the wet season has short rains (November - December) and long rains (March - May). The main vegetation type covering the area is acacia woodlands in the western and broad leaf forests in the northern (Estes et al., 2012; Mmassy et al., 2018).

3.3.1 Robanda Village Robanda village is located closer to the SNP for about 3 km (Figure 3.1). It is found in Serengeti district with a total population of 4,735 and 471 households. The village land is surrounded by protected areas such as SNP, Grumeti and Ikorongo Game Reserves and Ikona Wildlife Management Area (WMA). This village acts as a wildlife corridor, which is normally used by migratory animals, particularly wildebeest and zebra (Mwakatobe et al., 2012; Nyahongo et al., 2009). The majority of the local people belong to the Ikoma tribe, followed by immigrants who include Isenye, Ikizu, Zanaki and Kurya tribes.

The main economic activities include farming, pastoralism, agro-pastoralism, hunting, charcoal making, making local brew, small scale business and temporary employment in the adjacent protected areas. The main crops grown include maize, beans, millet (Panicum miliaceum), sorghum and vegetables. Livestock kept are cattle, goat, sheep, donkey (Equus africanus), pig (Sus domesticus), and poultry. Bushmeat is used as the main source of meat protein food in households and supplemented by livestock meat and fish, including sardines. This village has more diverse economic activities which generate income for the households. Being close to protected areas gives more opportunities for employment and business like selling of environmental products such as thatching grass, firewood and polls to tourist campsites (Kyando et al., 2019). It has also set aside 70% of the village land for

31 conservation of wildlife through Ikona WMA and the rest of the land is used according to village land use plan. Dispite the conservation awareness which is widely spread among villagers, some people still do the illegal harvest of natural resources mainly bushmeat.

3.3.2 Rwamkoma Village Rwamkoma Village is located 27 km from SNP boundary to the village center (Figure 3.1). It is located in Butiama district with a total population of 4,821 and 802 households. The main economic activities of local people are agricultural production (crop and livestock production) and small scale business. This village has good fertile soil suitable for agriculture and two rainfall seasons that enhance crop production with relatively large harvest. Agricultural production is the main source of income followed by small scale business. Other income sources such as employment in tourism activities and conservation areas, hunting and bushmeat trade have low contribution because the village is located a bit far from protected areas. The food crops cultivated in the area includes beans, sweet potatoes, maize, rice, cassava, sorghum, millet and vegetables. Livestock kept are cattle, goat, sheep, pig, and poultry. The majority of the local people belong to the Zanaki tribe and followed by Ikizu, Jita, Luo and Kurya tribes. Sardines (small dried fish) are used as a main source of animal protein food in households in addition to other fish types and livestock meat. Bushmeat is rarely consumed by households because it is sold at higher prices by bushmeat traders.

3.3.3 Kowak Village This village is located 58 km from the center of the village to the SNP boundary (Figure 3.1) and it is found in Rorya district. It has a total population of 4,382 people and 1,005 households. The main economic activities include farming, pastoralism, agro-pastoralism, fishing, making local brew and small scale business. The food crops cultivated in the area are sorghum, maize, cassava, millet and vegetables. Livestock kept are cattle, goat, sheep, donkey, pig, and poultry. Most of local people are of Luo tribe and a few Kurya ethnic groups. Sardines are used as the main source of animal protein food in households in addition to domestic meat and other fish

32 types. Bushmeat consumption by households is less frequently due to limited availability and higher prices.

3.4 Research Design and Sampling This study followed a longitudinal research design in which data were collected for the same subjects repeatedly at the interval of season (wet and dry) over the year. The design was adopted because of the seasonal influence on the availability of bushmeat in the study area which is associated with migratory herbivores particularly wildebeest and zebra (Nyahongo et al., 2009). The nature of the bushmeat hunting activities being illegal in western Serengeti was also among the reasons for this design.

3.4.1 Sampling Unit This study used household as a sample unit and a list of households was obtained from the village register in the office of the Village Executive Officer (VEO). A household (HH) was referred to a person or group of people living in the same compound, eating together and headed by the same household head (URT, 2013).

3.4.2 Sample Size The sample size was calculated based on the sampling frame (total number of households) from the study villages where at least 5% of all households is included. Within three selected villages, 50 households were selected randomly for household questionnaire survey in each village per season making a total of 150 households which represent 6.6% of the total households. Other households were selected by snow balling during observations and recording of bushmeat packages until the saturation point was attained (Sandelowski, 1995). Schoolchildren interviews comprised 40 class-four students selected from one primary school in each village.

3.4.3 Sampling procedures A list of households for questionnaire survey was randomly selected from the village register. The sampling technique used was simple random sampling. Each household was assigned a number and households whose number was produced by a random number generator in Excel, were selected for interview. Snowballing sampling was used to select respondents who participated in illegal hunting and bushmeat trade.

33 Research assistants from the village helped in identification of the first respondents who served as initial sample and a way to get others. This sampling technique was used in order to overcome a challenge of difficulties in finding respondents due to the illegal nature of the ctivities (Knapp, 2012; Nielsen, Jacobsen, et al., 2014).

3.5 Data Collection Methods Data were collected through household questionnaire survey, dietary recall survey, focused group discussion (FGD) and observations during both dry (September- October 2017) and wet (April-May 2018) seasons. In this study, both direct and indirect questioning techniques have been used to collect information on the contribution of bushmeat to households in western Serengeti. Direct interview or self reporting and dietary recall surveys were used to collect information on consumption frequencies of bushmeat and other meat types in households on weekly bases. Data on household meat consumption were collected based on the number of meals consumed because respondents were not conversant with the actual amount in kilogram since bushmeat was measured in pieces or packages.

It was also applied for other meat types since very few sellers had weighing scales and therefore they relied on estimates. This was observed during pilot study in questionnaire pre-testing and observations in local markets where fish were sold in bulk and bushmeat was sold in packages secretly. Bushmeat consumption records, is an indirect measure of illegal activities from people although it cannot give more information on how and where the meat was obtained (Knapp et al., 2010). In order to get such information, bushmeat hunters and traders were also interviewed through self reporting based on trust. However, respondents were selected non-randomly through snow balling techniques due to the sensitivity of the activity.

In order to overcome a challenge of adult strategic behaviour in sharing information about bushmeat, this study used both adults and schoolchildren to collect information on the frequency of bushmeat consumption at household level in the study villages. The information reported by schoolchildren was compared with that of adults to assess the accuracy of information provided. Moreover, the study also intended to collect information about the contribution of bushmeat income to household income

34 in western Serengeti. Such information was collected from bushmeat hunters and traders through a snowballing technique with the help of informers and local assistants from the study villages. An informed consent form was developed, and schoolchildren and adults, were all verbally explained the objectives of the study and that they could withdraw their participation at any time in the study. Obtaining children‟s agreement to participate in the study was made in the presence of their legal guardian.

A preliminary study was done for a period of two weeks before the actual data collection for three main activities; selection and training of field assistants, questionnaire pre-testing and participatory wealth ranking exercise. In each village, two field assistants were selected and trained for data collection. The selection of research assistants was done together with the VEO in each village. Local assistants were preferred in order to reduce bias and also to give interpretations for respondents who do not speak Kiswahili. Selected assistants were trained on how to approach respondents and ask question from the study questionnaire as well as recording of the information collected.

Questionnaire pre-testing was done in order to test if the questions were clear, relevant, and likely to generate adequate information as required by the study objectives. The questionnaires were pretested at Nyatwali village in Bunda district. After the exercise, editing was done for parts that required some modifications and the questionnaire was reformed and used in the actual data collection.

A participatory wealth ranking exercise was conducted through a focused group discussion with the help of VEO and local people selected from three different groups of adult male, adult female and youth both males and females. Each goup composed of eight participants and had separate discussion guided by the researcher (Plate 3.1). The group participants were identified by the researcher with the help of VEO. The wealth indicators were identified by the local people in a respective village within the FGD groups. Such indicators included the type of house, number and type of livestock owned and household cash income. Both cash income and assert wealth were used to assess wealth status of households, because poverty is

35 sometimes a transitory period if the household does not have cash income but have asserts (Nielsen, Pouliot, & Bakkegaard, 2012). The discussion agreed on wealth criteria for three wealth groups of poor, middle and rich households as presented in Table 3.1.

Table 3. 1: Wealth ranking criteria identified in the study villages Village Household wealth categories Poor Middle Rich Robanda No livestock or <10 Possess 10-50 cows Possess >50 cows cows Income > US$ 1 per Income >US$ 2 Income ≤ US$ 1 per day per day day Brick-walled, Brick-walled, Mud-walled, not cemented floor and cemented floor and cemented floor and iron-sheet house. corrugated iron- grass-thatched house. sheet house. Rwamkoma -No livestock or ≤ 5 Possess 10-30 cows Possess >30 cows cows Income > US$ 1 per Income >US$ 1 -Income ≤ US$ 1 per day per day day Brick-walled, Brick-walled, -Mud-walled, not cemented floor and cemented floor and cemented floor and corrugated iron- corrugated iron- grass-thatched house. sheet house. sheet house. Kowak No livestock or < 5 -Possess 5-10 cows Possess > 20 cows cows -Income > US$ 1 per Income >US$ 2 Income ≤ US$ 1 per day per day day -Brick-walled, Brick-walled, Mud-walled, not cemented floor and cemented floor and cemented floor and iron-sheet house. corrugated iron- grass-thatched house. sheet house.

36

Plate 3. 1: Group discussion for participatory wealth ranking exercise in Robanda village

3.5.1 Household Questionnaire Survey Semi-structured questionnaires were used to collect information from household heads who were selected randomly within the study villages (Appendix 1). Each household was assigned a number, and households whose number was produced by a random number generator in Excel, were selected for interview. The research team (both researcher and field assistants) administered household questionnaires face-to- face at the household area. This type of interview can help in assessing the accuracy of the information, keep focus to objectives, capture and control emotions and behaviours of the respondents. It also had the advantage of providing opportunity for probing and clarifying ambiguous questions (Kothari, 2004). The questions were structured as open ended and closed ended in order to give room for interviewees to express their views in details, also to simplify coding of the answers and to serve time for both parties. The questionnaire used was tested in the pilot area during preliminary studies.

3.5.2 Dietary Recall Survey A standard 24 hours dietary recall method (FAO, 2011) was used to collect information from schoolchildren on the number of meals eaten per day for a period of one month in each season. Daily dietary recall was done for a period of four weeks

37 in order to have sufficient information since too short periods as well as long periods were found to be sources of error (Knapp et al., 2010; Rentsch & Damon, 2013). Dietary recall questions reduce tension to respondents since the questions are designed in a way that bushmeat consumption can be reported together with other meat types consumed.

3.5.3 Focused Group Discussion Focus group discussion (FGD) refers to structured discussion which involves small groups of people with a facilitator who leads the discussion on specified topic using preset questions (Masadeh, 2012). FGDs are more useful when collecting public views on certain issue as it involves many participants at a time. These discussions can also help in understanding beliefs and experiencies of participants on the issues related to the research topic (Dawson, 2007).

As recommended by (Dawson, 2007), each FGD comprised a small group of people usually six to eight participants for effective FGD. In this study three FGDs per village were conducted, composed of eight participants selected from adult men, adult women and youth (both males and females). The discussion intended to collect information about bushmeat prices and availability, market place, hunting areas, hunting tools, ways of transportation and reasons for hunting.

3.5.4 Observations and Recording of Bushmeat Packages Bushmeat packages were observed and recorded from households in the study villages. This study used participant observation in order for the researcher to be familiar with the local activities in the study villages and also to build trust with the local people (Dawson, 2007; Kothari, 2004). Local assistants were used to connect the researcher with hunters and bushmeat traders which helped the researcher to observe some of the bushmeat packages in their households. This was important since the researcher need to get access to the community studied in order collect information under participant observation method (Dawson, 2007). The researcher spent two weeks in each village per season observing the local communities‟ daily activities and food sources consumed.

38 The identification of bushmeat species from bushmeat packages found was done by local assistants who also helped to cross-check if the respondents‟ identifications were true. Local informers helped to locate hunters and bushmeat traders as well as bushmeat packages available in the village. Local assistants helped to record data on bushmeat packages found in households of hunters and bushmeat traders. Information recorded includes number of packages, state of the meat (fresh or dried), bushmeat species name, source (hunted or purchased), price per piece, hunting area and respondent type (hunter or trader) (Appendix 4).

3.6 Data Collection Data were collected from different groups of respondents including schoolchildren and adult respondents selected from groups of hunters, bushmeat traders and regular households i.e those not involved in hunting and trading bushmeat. More information about data collection from each group of respondent is explained below.

3.6.1 Data Collection from Schoolchildren Approval to conduct the study was obtained following an ethical evaluation by the National Health Research Ethics Committee under the National Institute for Medical Research (NIMR) with reference number (Ref. NIMR/HQ/R.8a/Vol.IX/2609) (Appendix 6). Schoolchildren were interviewed to provide relevant information on frequency of bushmeat consumption by households. Data were collected through interviews during the dry season in September-October 2017 and the wet season in April-May 2018. Collecting data in seasons was done in consideration of the high influence of seasonality and the wildlife migration on the availability of bushmeat in the study area (Nyahongo et al., 2009).

Interviews were conducted with schoolchildren from primary schools in the study villages. Forty schoolchildren were selected randomly from the standard four class in each school. This age group represented children between 9 and 12 years, which means that they were old enough to recall and explain what they had consumed (Diep et al., 2015), but too young to have participated in hunting, which may increase awareness about the illegality and possible sanctions for hunting illegally.

39 Previous studies on bushmeat consumption have also used schoolchildren (Haule et al., 2002; van Vliet et al., 2015).

Plate 3. 2: The researcher discussing with schoolchildren in Robanda primary school

On the first day of the data collection in each school, the research team discussed general issues like health, nutrition and biology with the students to establish a good rapport with the children (Plate 3.2). Subsequently, the team implemented a questionnaire enquiring about pupils‟ household socio-demographic characteristics with the help of the teachers. The questions were explained in plenum and schoolchildren were given time to complete the questionnaire with the help of the teacher and researchers. The questionnaire obtained data about the children (age and gender) and their families (number of household members and household head occupation), and meals consumed over the 24 hours in breakfast, lunch and dinner.

On the second day of the study, schoolchildren were asked specifically about types of food including meat types consumed in meals in their household the past 24 hours. This questionnaire was repeated each day of the subsequent four consecutive weeks (data for weekends were recorded on the following Mondays) in each season. The

40 final combined sample contains data about 21,336 meals for 127 pupils from the three villages after attrition.

3.6.2 Data Collection from Adults Data on bushmeat consumption frequency from adults was collected through household questionnaire surveys conducted in the same three villages over the same period as the data from schoolchildren. Households were selected randomly and the head of the household was interviewed. The questionnaire used contained questions about household demographic data and socioeconomic information such as age, gender, education, occupation and income sources. Although the target was to interview household heads, other members aged 18 years and above were also selected when the household head was absent (Plate 3.3).

Plate 3. 3: The researcher interviewing a respondent in Kowak village

The questionnaire also inquired about the frequency of meat containing meals consumed the week before the interview. Incase the household head fail to recall and ask for household members to help, the wife or female members were invited to give

41 such informations since they were the onces who prepare food in many households. Overall, 150 respondents selected randomly from the three study villages were interviewed in both seasons.

Data on the role of bushmeat to household income was collected through household questionnaire surveys (Appendix 2) and observations from hunters and bushmeat traders conducted in September-October 2017 and April-May 2018. Snowballing was used to locate hunters and bushmeat traders, and local assistants together with local informers helped to locate bushmeat packages within the study villages. Local assistants supervised by the researcher, helped identify the first respondent and also recorded bushmeat packages observed. In snowball sampling, once a potential respondent is identified, he or she is asked to name others who might also be potential for the study. Incase where the chain broke, the local assistants helped to identify next respondent and the chain continue until the saturation point where the named respondents had already being interviewed. The interviews were done in their households with higher confidentiality. Overall, 96 households were interviewed in all three villages and found to have hunters and bushmeat traders. The value of bushmeat possesed was estimated by multiplying the number of packages by the current bushmeat price. The prices were identified for both fresh and dried meat across seasons in all villages based on the information provided by the respondents since the bushmeat market was conducted secretly. Other socioeconomic data and information on the assets owned by the household were also collected.

Respondents were also asked about income sources including bushmeat income on monthly bases. All income sources both cash and subsistence income identified by each household were recorded and estimated in their monetary value. The total household income was estimated from summation of all income sources. Household income sources identified were agricultural income, non-agricultural income (employment and business, remittance and pensions) and bushmeat income. Household dependence on bushmeat income was calculated as the percentage contribution to total household income.

42 Cash income refers to income earned from business and subsistence income means income generated from own production and consumption. Cash income obtained from products sold by the household was estimate by multiplying the amount of a product and the average market price (Appendix 7), while subsistence income was estimated by multiplying the quantity of the product harvested and the average market price. Calculated incomes are net income, i.e., the gross value of products minus costs of purchased inputs but not excluding own-labour. Calculated values were converted to US$ based on the average annual exchange rate of 1 US$ = TZS 2,250 for 2018.

3.7 Data Analysis The average frequency of consumption (i.e., times per week) in each village was calculated for each animal meat food source in each season by dividing the average number of meat meals of each type by the number of days in the recording period. Other meat types were grouped into two main categories; “domestic” for all livestock meat types including poultry and “fish” for all species and types of fish including sardines. Shapiro-Wilkes tests were used to test the assumption of normality (Zar, 2010) and the results showed that most of data were not normally distributed.

The Kruskal-Wallis test was used to test the differences in bushmeat consumption between villages and the Dunn‟s post hock test was used to test the significance of the differences. Wilcoxon tests were used to compare seasonal variations and stated bushmeat consumption frequencies between schoolchildren and adults in the same village. As stated bushmeat consumption frequencies for adults was obtained using recall during the first week of the survey in each village, the first week of data collected from the schoolchildren was selected for comparison. This approach was adopted in order to minimise bias induced by temporal variations in bushmeat consumption.

The factors influencing bushmeat consumption were evaluated by the Generalised Linear Model (GLM) with logarithmic transformation and standard specification of quasi-Poisson family and a canonical link function (Myers, Montgomery, Vining, & Robinson, 2010) since the data were not normally distributed. Variables selected for the model test were derived from previous studies related to this study as well as the

43 general economic theory (Fischer et al., 2014; Knapp et al., 2010; Nyahongo et al., 2009). The variables were distance to protected area boundary (in km), season (wet or dry season), age of respondent (child or adult), household size, and occupation of the household head (peasant and pastoralist vs employed or managing a small-scale business) (Table 3.2). Evaluation of the model was done using the dispersion parameter and variance inflation factor (VIF) (Lindsey, 2008). The VIF was calculated to detect multicollinearity.

44 Table 3. 2: Explanatory variables and expected sign of coefficient for variables selected as indicators for testing hypotheses about the determinants of frequency of reporting bushmeat consumption in the western Serengeti Covariate Unit Sign Hypothesis Distance Km - Bushmeat consumption frequency is inversely related to distance to the PA boundary as an indicator of availability and price of bushmeat. Seasonality Dummy variable + Bushmeat consumption is higher (Dry season coded in the dry season than in the wet as 1 and wet season season due to the presence of the as 0) wildebeest migration in the dry season. Age group Dummy variable + Children state higher bushmeat (Children coded as 1 consumption frequency than adults and adults as 0) due to lack of concern about reprisal as a consequence of the illegality of bushmeat hunting. Household Number of people + Larger households are more size efficient in wealth generation and therefore can afford to purchase more bushmeat and allocate household members to hunting. Household Dummy variable + Bushmeat consumption frequency head (Employed/business is positively associated with formal occupation coded as 1 and occupation or running a small-scale peasant/pastoralist as business as an indicator of wealth. 0) Domestic Frequency - Bushmeat consumption frequency meat is inversely related to domestic consumption meat consumption frequency as an indicator of wealth and fulfilment of protein needs from other sources. Fish Frequency - Bushmeat consumption frequency consumption is inversely related to fish consumption frequency as an indicator of wealth and fulfilment of protein needs from other sources.

45 The Kruskal-Wallis test was used to test the variation in household dependence on bushmeat income and the Dunn‟s post-hock test was used to evaluate the differences among villages. Since most of the hunters were also traders therefore, both were grouped together in analysis of comparison with regular households and were tested by the Wilcoxon test. All statistical tests used were selected after Shapiro-Wilkes test results on normality assumption. Heckman sample selection model was used to evaluate predictors of household participation in hunting and bushmeat income reliance (Bakkegaard, Nielsen, & Thorsen, 2016; Toomet & Henningsen, 2008). This model is useful in overcoming challenges observed in other regression models including selection bias when households self-select into the activity studied (Certo, Busenbark, Woo, & Semadeni, 2016). The Heckman model can estimate consecutively the predictors of household participation in hunting and also the factors influencing reliance on such activities. The model has two main equations where the first help to estimate the household selection into the activity and the second part give the outcome. The Heckman model assumptions help to overcome problems associated with endogeneity which is common in other selection models (Toomet & Henningsen, 2008). The selection equation is defined as: w*ᵢ = γꞋZᵢ + uᵢ (1) where, w*ᵢ is the latent variable, related to a set of exogenous variables, Zᵢ, and where wᵢ = 1 if wᵢ > 0 and wᵢ = 0 otherwise. The probability of observing participation, i.e. wᵢ = 1, as a function of Zᵢ is defined by a probit model: Prob(wᵢ = 1|Zᵢ) = ɸ(γꞋZᵢ) (2) Prob(wᵢ = 0|Zᵢ) = 1 - ɸ(γꞋZᵢ) (3)

When wᵢ = 1, bushmeat income is observed as outcome reflected in bushmeat reliance above zero for household i, namely yᵢ. The outcome part of the model will then describe the outcome in terms of the reliance on bushmeat income, and its relation to a subset of variables X (which may overlap with Z):

yᵢ = βꞋxᵢ + ɛᵢ, where wᵢ = 1 (4)

The error terms are assumed to be distributed as (uᵢ, ɛᵢ) * bivariate normal [0, 0, 1, σɛ, ρ], allowing for possible correlations in the error terms.

46 Variables tested by the model were distance to the PA boundary (in km), household income per capita, household size, actor group (hunters or bushmeat traders) and the age and gender of the household head (Table 3.3). All statistical tests were done in R- Studio (Version 1.1.456) using a significant level of P ≤ 0.05.

47 Table 3. 3: Explanatory variables and an expected sign of the coefficient in the Heckman model testing hypotheses about the likelihood of participation in hunting and trading bushmeat and its outcome as the magnitude of reliance on bushmeat income Variable Unit Expected sign Hypothesis Participation Reliance Distance to Km Households are more likely to the PA hunt/trade and rely more on boundary bushmeat income the closer - - they are to the boundary due to lower opportunity costs and higher availability. Household Cash income Households with higher cash from all income from other sources income household are less likely to hunt/trade - - income and relies less on bushmeat sources in income due to more US$ remunerative alternatives. Gender of Male-headed Female-headed households the =1 or are less likely to contain household female- hunters/traders and hence - - head headed = 0 relies less on bushmeat income due to lower skills and labour availability. Age of the Years Households with older heads household are more likely to hunt/trade head and rely more on bushmeat + + income than households with younger heads due to fewer alternatives. Household Number of Larger households have more size people in the excess labour and are more household likely to contain + - hunters/traders but are also more efficient in wealth generation and therefore rely less on bushmeat income. Actor Hunter = 1 Hunting households rely more group or trader = 2 NA + on bushmeat income than traders.

48 3.8 Validity and Reliability 3.8.1 Validity Validity refers to the accuracy of an assessment measured by a researcher (Heale & Twycross, 2015). The validity of this study was enhanced by different measures employed during the research process. The triangulation method on which data collection involved different methods such as questionnaire survey, dietary recall survey, FGD and observations) in order to increase the validity of the information obtained. Furthermore, the study involved adoption of the random sampling of households for questionnaires survey for adult respondents which was compared with the information collected from schoolchildren on bushmeat consumption frequencies. Also the use of the local research assistants from the study villages who were trained before data collection activities helped to increase the validity of the information collected. Moreover, the questionnaire used was pre-tested during preliminary study in order to assess the relevance of the questions in answering research objectives. All these approaches were intended to increase the validity of the information obtained from different angles so as to meet the study objectives.

3.8.2 Reliability Reliability in research activities means consistency of the activities involved in a research process. It is a measure of quality in quantitative studies on the accuracy of research instruments used (Heale & Twycross, 2015). This study used a longitudinal approach and data were collected twice during the dry and wet season using the same procedures in the same study villages so as to enhance reliability of the information obtained. Moreover, data were collected from schoolchildren as well as from adults in order to reduce bias since adult respondents are more sensitive to illegal bushmeat hunting activities. Reliability of the information obtained was also enhanced by the use of standard methods and instruments (GPS, weigh balance, questionnaire and checklist) in data collection.

3.8.3 Ethical Considerations The directorate of graduate studies at the University of Dodoma approved the study in order to fulfil the requirements for PhD studies. Permission to conduct the study in villages around the SNP was obtained from the respective district authorities

49 following the research permit authorized by TAWIRI, COSTECH and NIMR. The study intends to protect the privacy of people involved and avoid them being harmed physically or mentally. Respondents were free to participate in the study on their will with no prior force. The information collected was treated confidential due to the illegal nature of bushmeat hunting activities and only used for academic purpose.

3.9 Delimitations and Limitations of the Study Delimitations refer to the scope of the study selected in order to meet the objectives. It helps in planning research activities which are manageable and achievable depending on the resources available. Normally research studies have delimitations due to constrains in time and fund in addition to others. Study limitatations refer to the challenges encountered in the study mainly during research activities. This study had delimitations as well as some limitations due to various reasons explained below.

3.9.1 Delimitations of the Study The study focused mainly on bushmeat as an ecosystem good provided by the GSE to local communities living around. This study was based on provisioning services which include bushmeat as one of the food sources provided by the ecosystem. The main aim was to explore the role of bushmeat as food and the contribution of bushmeat income to household income in western Serengeti. Other ecosystem services provided by the GSE were not considered and also the eastern and southern parts of GSE were not included in this study because of limitations in time and resource. However, an attempt was made to cover more than three villages in order to get more samples, but it was not possible due to the same reasons. Despite such delimitations, the study objectives were achieved since the selected villages represented the local communities in western Serengeti.

3.9.2 Limitations of the Study The study encountered some limitations and most of them occurred during data collection. The main challenge was to ask information about bushmeat hunting activities because bushmeat was illegally obtained. Moroever, it was also a challenge to observe the bushmeat trade which was operating in a secrete market, with meat sold within households door to door. Bushmeat was sold mainly to local people and

50 only to known visitors whom they trusted. The identification of bushmeat packages was done by using local knowledge. Further identification using molecular techniques was limited by resources availability. There was also a language barrier in some households, where the respondents did not speak Kiswahili. In order to overcome all these challenges, the study employed defferent data collection methods and techniques including participant observations and also the use of local research assistants from the study villages. However, the information reported by respondents can have some degree of certainty but still be useful for further planning and development of strategies to reduce the problem of illegal bushmeat hunting activities.

51 CHAPTER FOUR

RESULTS

4.1 Overview In this chapter, a detailed reporting of the study results is presented. Findings are generally organised around the research objectives. The results are grouped into sections which present results from both descriptive and inferential statistics.Tables, figures and plates were also used to summarise the information. Some of these results have already being published in journal articles. First paper was published in May 2019 in the International Journal of Biodiversity and Conservation; volume 11, issue number 5 and the second paper in July 2019 in the Environment and Natural Resources Research Journal; volume 9, issue number 3.

4.2 Demographic and socioeconomic characteristics of the respondents The general characteristics of the respondents both schoolchildren (n = 127) and adults (n = 150) from regular households and the sample percentage composition in each village are summarised in Table 4.1 and 4.2. The majority of the schoolchildren (88.5%) were at the age between 9-12 years and few (11.5%) had 13-14 years. Also the overall proportion of girls (52.8%) was relatively higher than boys (47.2%). The number of household members (household size) reported by schoolchildren were significantly different with those reported by adult respondents (Wilcoxon test; H = 8203.5; P = 0.045). Overall average household size reported by both schoolchildren and adult respondents was seven people while children reported more households (53.5%) with less than seven members and adults reported more households (52%) with more than seven people (Table 4.1 and 4.2).

The overall adult sample (n = 150) from regular households composed of more females (60%) than males (40%). Most of the respondents (48%) were at the age between 29-48 years followed by 49 and above years (34%) and 21-28 years (18%). The majority (86.6%) of the respondents had primary education and few (9.8%) had higher levels of secondary education and college and (3.7%) had no formal education. Agricultural production was the main source of income for most households (88.7%) and small scale farming (62.7%) was the leading followed by

52 livestock keeping (26%) and business and wages income (11.3%). Most of the households (59.3%) were economically poor with few middle earners (27.3%) and rich (13.3%) households (Table 4.2).

Table 4. 1: Baseline information of the respondents (schoolchildren; n = 127) Variable Study villages Village characteristics Robanda Rwamkoma Kowak Total village population 4,735 4,821 4,382 Number of households 471 802 1,005 Distance to PA (Euclidean distance) (km) 3 27 58 Schoolchildren interviewed 46 49 32 Gender Girls (%) 52.2 57.1 46.9 Boys (%) 47.8 42.9 53.1 Age 9-12 years (%) 91.3 83.7 90.6 13-14 years (%) 8.7 16.3 9.4 Household size Less than 7 people (%) 80.4 42.9 59.4 More than 7 people (%) 19.6 57.1 40.6 Household heads occupation Peasants (%) 26.1 87.8 87.5 Pastoralists (%) 32.6 0 0 Employment and Small scale business (%) 41.3 12.2 12.5

53 Table 4. 2: Baseline information of the adult respondents from regular households (n = 150) Variable Study villages Village characteristics Robanda Rwamkoma Kowak Total village population 4,735 4,821 4,382 Number of households 471 802 1,005 Distance to PA (Euclidean distance) (km) 3 27 58 Number of respondents interviewed 49 49 52 Gender Males (%) 26.5 49 42.3 Females (%) 73.5 51 57.7 Age 21-28 (%) 38.8 2 13.5 29-48 (%) 34.7 47 61.5 49 and above (%) 26.5 51 25 Household size Less than 7 people (%) 65.3 42.9 36.5 More than 7 people (%) 34.7 57.1 63.5 Education level Primary (%) 77.6 93.9 88.4 Above primary (%) 18.4 4.1 3.9 Illiterate (%) 4 2 7.7 Household heads occupation Peasants (%) 42.9 81.6 65.4 Pastoralists (%) 34.7 18.4 25.0 Employment and Small-scale business (%) 22.4 0.0 9.6 Household wealth rank Poor (%) 53 67.4 57.7 Middle (%) 24.5 28.6 28.9 Rich (%) 22.5 4 13.4

54 4.3 Seasonal variation in frequency of bushmeat consumption by households Bushmeat meals were more frequently consumed during the dry season 66% (n = 277) than during wet season 34% (n = 277). Altogether 572 meals contained bushmeat out of 3,623 meat meals recorded in the overall sample from both adults and schoolchildren (n = 277). In total, bushmeat constituted 15.8% (n = 277) of meat containing meals. The meat of domestic animals constituted 18% while fish was by far the most common source of animal protein at 66.2% of meals containing meat. More bushmeat was significantly consumed in meals for dinner 55.9% (n = 277) than in meals for lunch 44.1% (n = 277) (Wilcoxon test; W = 14,836; P < 0.001) and no bushmeat was consumed during breakfast.

The variation in childrens‟ bushmeat consumption frequency between weeks was tested in each season, and no significant difference between the four weeks was found (Kruskal-Wallis test; dry season, H = 3.18; P = 0.364; wet season, H = 1.32; P = 0.725). A combined analysis of data from both adults and schoolchildren (n = 277) shows that, the average number of bushmeat meals per week was significantly higher in the dry season (1.4 ± 0.12) than in the wet season (0.7  0.07), (Figure 4.1; Wilcoxon test; W = 42.50; P = 0.018). In the closest village (Robanda) where bushmeat was most frequently consumed (96.3% of meat containing meals), the consumption frequency was also significantly higher during the dry season than in the wet season (Figure 4.1; Wilcoxon test; W = 7,137.50; P < 0.01).

55

Figure 4. 1: Average number of bushmeat meals consumed per week reported by adults and schoolchildren in the wet and the dry season in Robanda, Rwamkoma and Kowak at increasing distance from the PA boundary.

4.4 Contribution of bushmeat to household meat consumption as a function of distance from the SNP boundary Bushmeat consumption differed between study villages (Kruskal-Wallis test; H = 454.2; P < 0.001) and was consumed very frequently in Robanda the closest villages to SNP at 96.3% (n = 95) of meat meals, while very little bushmeat was consumed in the intermediate village, Rwamkoma 1.4% (n = 98) and distant village, Kowak 2.3% (n = 84). Bushmeat consumption was negatively correlated with fish consumption (Spearman rank correlation test; r=- 0.58; P < 0.001) and more fish meals were consumed in the intermediate village, Rwamkoma and in the distant village, Kowak than in Robanda, the closest village to SNP (Figure 4.3; Kruskal-Wallis tests; H = 262.69; P < 0.0001).

Bushmeat consumption frequencies were significantly higher in the closest village (Robanda) than in the intermediate village, Rwamkoma and in the distant village, Kowak in both seasons (Figure 4.1; Kruskal-Wallis tests; H = 219.88; P < 0.001 and H = 240.57; P < 0.001, in the wet and dry season, respectively). Furthermore, a post

56 hoc Dunn‟s test shows a significantly higher consumption in Robanda than in Kowak (Z = 12.28; P < 0.001 and Z = 12.89; P < 0.001 in the dry and wet season, respectively), as well as in Rwamkoma (Z = 13.23; P < 0.001 and Z = 13.79; P < 0.001, in the dry and wet season, respectively). However, the bushmeat consumption frequency did not differ significantly between Rwamkoma and Kowak in either season (Z = 0.37; P = 0.714 and Z = 0.28; P = 0.78, in the dry and wet season, respectively).

4.4.1 Contribution of other meat types to household meat consumption as a function of distance from the SNP boundary Overall, the consumption frequency of other meat types (domestic meat and fish) was significantly different between study villages (Kruskal-Wallis tests; Domestic meat: H = 16.65; P < 0.001 and fish: H = 154; P < 0.001; Figure 4.2 and 4.3, respectively). The consumption frequency of domestic meat was significantly lower in the distant village, Kowak than in the closest village, Robanda (Dunn‟s tests; Z = 3.9; P < 0.001) and in the intermediate village, Rwamkoma (Z = 3.11; P = 0.004). However, the difference in the consumption of domestic meat between Robanda and Rwamkoma was insignificant (Dunn‟s test; Z = -0.853; P = 0.39).

The consumption frequency of fish was significantly lower in the closest village, Robanda than in the intermediate village, Rwamkoma (Dunn‟s tests; Z = 10.4; P < 0.001) and in the distant village, Kowak (Z = -11; P < 0.001). However, the difference in the consumption frequency of fish between Rwamkoma and Kowak was insignificant (Dunn‟s test; Z = -1.07; P = 0.29).

During dry season, the consumption of other meat types also differed significantly between villages (Kruskal-Wallis tests; Domestic meat: H = 17.23; P < 0.001 and fish: H = 79.04; P < 0.001). The consumption frequency of domestic meat was significantly higher in the closest village, Robanda than in the intermediate village, Rwamkoma (Dunn‟s tests; Z = -3.285; P = 0.002) and distant village, Kowak (Z = 3.81; P < 0.001) whereas, the difference between Rwamkoma and Kowak was insiginificant (Z = 0.68; P = 0.498).

57 Furthermore, a post hoc Dunn‟s test shows that fish meals were less consumed in the closest village, Robanda than in the intermediate village, Rwamkoma (Z = 7.8; P < 0.001 and in the distant village, Kowak (Z = -7.4; P < 0.001) during the dry season. However, fish consumption frequencies in the intermediate village, Rwamkoma and in the distant village, Kowak did not differ significantly (Dunn‟s test; Z = 0.15; P = 0.88).

Figure 4. 2: Average number of domestic meat meals consumed per week reported by adults and schoolchildren in the wet and dry season in Robanda, Rwamkoma and Kowak at increasing distance from the PA boundary.

During wet season, the consumption of domestic meat and fish also differed significantly between study villages (Kruskal-Wallis tests; Domestic meat: H = 14.78; P < 0.001 and fish: H = 76.89; P < 0.001). The consumption frequency of domestic meat was significantly lower in the intermediate village, Rwamkoma than in Robanda (Dunn‟s tests; Z = 2.25; P = 0.049) and Kowak (Z = 3.81; P = 0.0004) whereas the difference between Robanda and Kowak was insignificant (Z = 1.63; P = 0.1).

58 A post hoc Dunn‟s test shows that fish meals were less consumed in the closest village, Robanda than in the intermediate village, Rwamkoma (Z = 6.81; P < 0.001 and in the distant village, Kowak (Z = -8.14; P < 0.001) during the wet season. However, fish consumption frequencies in the intermediate village, Rwamkoma and in the distant village, Kowak did not differ significantly (Dunn‟s test; Z = -1.64; P = 0.1).

Figure 4. 3: Average number of fish meals consumed per week reported by adults and schoolchildren in the wet and dry season in Robanda, Rwamkoma and Kowak at increasing distance from the PA boundary.

4.5 Variation in the frequency of bushmeat consumption reported by adults and Schoolchildren Overall, schoolchildren reported a significant higher bushmeat consumption frequency on average (1.3 ± 0.18) than adults (0.8 ± 0.11) (Wilcoxon test; W = 33,526; P = 0.003). Furthermore, bushmeat consumption frequencies reported by both schoolchildren and adult respondents were significantly different between seasons (Figure 4.1). Schoolchildren reported a significantly higher bushmeat

59 consumption frequency (1.8 ± 0.22) than adults (0.9 ± 0.13) during the dry season (Wilcoxon test; W = 7,820.50; P = 0.002). However, during the wet season, the consumption frequencies of adults (0.6 ± 0.08) and schoolchildren (0.9 ± 0.12) were not significantly different (Wilcoxon test; W = 8,880; P = 0.21).

Furthermore, the overall bushmeat consumption frequency reported by schoolchildren was significantly higher (57.3%, n = 46) than adults (42.7%, n = 49) in the closest village (Robanda) where bushmeat was most frequently consumed (96.3%, n = 95) (Wilcoxon test; W = 3,155; P < 0.001). Moreover, schoolchildren (n = 46) (4.5 ± 0.3 and 2.3 ± 0.2, dry and wet season, respectively) reported significantly higher bushmeat consumption than adults (n = 49) (3 ± 0.2 and 2 ± 0.12, dry and wet season, respectively) in both seasons (Figure 4.1; dry season; Wilcoxon tests; W = 610; P < 0.001 and wet season; W = 840; P = 0.018).

In the intermediate (Rwamkoma) and distant (Kowak) villages, only few respondents (18.7%, n = 182) both adults and schoolchildren reported bushmeat consumption in their households and the difference between adults and schoolchildren was also significant (Figure 4.1; Rwamkoma; Wilcoxon tests; W = 4,500; P = 0.004 and Kowak; W = 2,652; P < 0.001). During dry season, bushmeat consumption frequency reported by schoolchildren was significantly higher than adults in the intermediate and distant villages (Rwamkoma; Wilcoxon tests; W = 1,050; P = 0.006 and Kowak; W = 586.5; P < 0.001).

In the wet season, schoolchildren reported more bushmeat consumption than adults in the distant village, Kowak (Wilcoxon test; W = 739.5; P = 0.028) whereas in the intermediate village Rwamkoma, bushmeat consumption frequency reported by schoolchildren did not differ significantly with that of adults (Wilcoxon test; W = 1,200; P = 0.32).

60 4.5.1 Variations in the consumption frequency of other meat types reported by adults and schoolchildren During dry season, the average consumption frequencies for both domestic meat and fish reported by schoolchildren were significantly higher than that of adults (Figure 4.2 and 4.3) (Domestic meat: Wilcoxon tests; W = 6302; P < 0.001 and Fish: W = 5395.5; P < 0.001). The consumption frequency of domestic meat differed significantly between villages (Figure 4.2) only for schoolchildren (Kruskal-Wallis tests; H = 10.45; P = 0.005) but not adults (H = 5.65; P = 0.056). Schoolchildren in Robanda reported higher domestic meat consumption than schoolchildren in Rwamkoma (Dunn‟s tests; Z = 2.87; P = 0.012) and Kowak (Z = 2.64; P = 0.017) while the difference between Rwamkoma and Kowak was not significant (Z = 0.081; P = 0.94).

The consumption frequency of fish meals also differed significantly between study villages (Figure 4.3) for both adults (Kruskal-Wallis tests; H = 32.72; P < 0.0001) and schoolchildren (H = 60.98; P < 0.001) during the dry season. A post hoc Dunn‟s test shows that both adults and schoolchildren in the closest village, Robanda consumed less fish than in Rwamkoma (Adults: Z = 4.2; P < 0.001 and Schoolchildren: Z = 6.73; P < 0.001) and in Kowak (Adults: Z = 5.47; P < 0.001 and Schoolchildren: Z = 6.61; P < 0.001). However, fish consumption frequencies in the intermediate village, Rwamkoma and in the distant village, Kowak did not differ significantly for either adults (Dunn‟s tests; Z = 1.25; P = 0.21) or schoolchildren (Z = 0.62; P = 0.54).

During wet season, the variation in the consumption frequency of domestic meat was statistically different between study villages only for schoolchildren (Kruskal-Wallis tests; H = 8.86, P = 0.012) whereas the consumption frequency reported by adults did not differ significantly between study villages (H = 5.26, P = 0.072). The differences were significant only between Rwamkoma and Kowak (Dunn‟s tests; Z = 2.97; P = 0.009) but not between Robanda and Rwamkoma (Z = 1.49; P = 0.135) or Robanda and Kowak (Z = 1.60; P = 0.22).

61 The consumption frequency of fish was significantly higher in the intermediate and in the distant villages for both adults (Kruskal-Wallis tests; H = 42.18; P < 0.001) and schoolchildren (H = 68.47; P < 0.001) in the wet season. Furthermore, a post hoc Dunn‟s test shows that fish consumption reported by schoolchildren differed significantly between the three villages (Robanda vs Rwamkoma: Z = 5.58; P<0.001, Robanda vs Kowak: Z = 7.99; P < 0.001 and Rwamkoma vs Kowak: Z = 3.06; P < 0.002). However, fish consumption reported by adults was lower in Robanda than in Rwamkoma (Dunn‟s tests; Z = 4.46; P < 0.001) and Kowak (Z = 6.32; P < 0.001), while consumption frequencies in Rwamkoma and Kowak did not differ statistically (Z = 1.85; P = 0.06).

4.6 Factors influencing bushmeat consumption in households The generalised linear model (GLM) revealed that, reported bushmeat consumption frequency is associated with the respondent category being positive and significantly associated with schoolchildren as informers. The frequency of bushmeat consumption is also significantly associated with the frequency of consumption of other meat types decreasing when the consumption frequency of domestic meat and fish increases. Other significant factors include a positive association with season and a negative association with distance from the PA boundary. Hence, bushmeat is more frequently consumed in the dry season but less frequently consumed as distance to the PA boundary increases. Other socioeconomic factors such as household size and occupation were not significantly associated with bushmeat consumption frequency (Table 4.3). The average variance inflation factor (VIF) was 1.11 indicating that multicollinearity is negligible and the dispersion parameter was 0.63 indicating that the model has no sign of overdispersion which can leads to type I error.

62 Table 4. 3: Regression coefficients of the quasi-Poisson Generalised Linear Model for predictors of household bushmeat consumption frequency Variable Estimate SE T P Intercept 1.58589 0.14036 11.299 <0.001 Age 0.65730 0.07050 9.323 <0.001 Domestic meat consumption frequency -0.20436 0.02974 -6.871 <0.001 (times per week) Fish consumption frequency (times per -0.14485 0.02462 -5.884 <0.001 week) Season 0.76243 0.07215 10.567 <0.001 Distance to PA boundary (27 km) -4.06938 0.28980 -14.04 <0.001 Distance to PA boundary (58 km) -3.29956 0.23122 -14.27 <0.001 Household size 0.02402 0.01629 1.475 0.141 Household heads occupation 0.06196 0.07680 -0.807 0.420 Model properties: Observations, 554; Respondents, 277; Dispersion parameter, 0.63

4.7 Demographic and socio-economic characteristics of adult respondents The composition and characteristics of adult respondents both hunters/bushmeat traders and regular households are presented in Table 4.4. The sample characteristics differed significantly between respondents in terms of the age and gender of the household head, occupation (in the category “others”) and distance to the PA boundary. Other characteristics such as household size, education level and household wealth rank were not statistically different (Table 4.4).

63 Table 4. 4: Percentage composition of the adult sample and comparison of means between hunters/bushmeat traders and regular households Variable (%) Hunters and Regular households Z P Traders (n = 96) (n = 150) Distance to PA boundary (km) 3 (Robanda) 79.2 32.7 27 (Rwamkoma) 11.5 30.7 -4.310 <0.001 58 (Kowak) 9.3 36.6 -4.677 <0.001 Gender Male 76 40 Female 24 60 -4.270 <0.001 Age 21-28 10.4 18 29-48 68.8 48 2.891 0.004 49 and above 20.8 34 1.819 0.069 Education level Primary 86.5 86.7 Above primary 11.5 8.7 0.395 0.693 Illitrate 2 4.6 -1.470 0.141 Occupation Peasants 36.4 62.7 Pastoralists 24 26 1.666 0.096 Others 39.6 11.3 2.898 0.004 Household wealth Poor 67.7 59.3 Middle 24 27.3 1.119 0.263 Rich 8.3 13.3 -0.643 0.520 Household size Less than 7 people 67.7 48 0.938 0.348 More than 7 people 32.3 52

64 4.8 Contribution of bushmeat to household income and the influence of distance from PA boundary on bushmeat hunting activities Household participation in bushmeat hunting activities and reliance on bushmeat income were significantly influenced by distance from the village to the PA boundary (Table 4.4 and Figure 4.4 and 4.5). Bushmeat income and reliance on bushmeat income were significantly higher in the closest village, Robanda compared to the intermediate village, Rwamkoma and distant village, Kowak (bushmeat income; Figure 4.4; Kruskal-Wallis tests; H = 24.025; P < 0.001 and bushmeat reliance; Figure 4.5; H = 24.789; P < 0.001). A post-hoc Dunn‟s test reveals significant differences in bushmeat income between Robanda and Rwamkoma (Z = - 4.315; P < 0.001) as well as between Robanda and Kowak (Z = 2.966; P = 0.006), while the difference between Rwamkoma and Kowak was not significant (Z = - 0.771; P = 0.441).

Figure 4. 4: Average bushmeat, livestock, crop and wage and business income of hunter and bushmeat trader households at increasing distance from the PA boundary.

65 Furthermore, similar differences were also observed in reliance on bushmeat income (Figure 4.5) where the differences between Robanda and Rwamkoma (Dunn‟s tests; Z = -4.009; P < 0.001) as well as between Robanda and Kowak (Z = 3.264; P = 0.002) were significant, while the difference between Rwamkoma and Kowak was not significant (Z = -0.318; P = 0.751). The average income earned from bushmeat was significantly higher for bushmeat traders (231.63 ± 12.47) than hunters (146.32 ± 7.44) (Wilcoxon test; W = 830; P = 0.021). Most hunters (72.22%) claimed to catch only one animal per hunting trip, and the majority (58.33%) claimed to hunt only one time per month on average during both seasons.

Figure 4. 5: Reliance (percentage contribution to total household income) of bushmeat, livestock, crop and wage and business income of hunter and bushmeat trader households at increasing distance from the PA boundary.

66 4.8.1 Bushmeat hunting activities and bushmeat packages observed and recorded A total of 149 bushmeat packages were recorded in possession of hunters and bushmeat traders (n = 96) in the study villages and some of the pictures are presented in Plate 4.1-4.4. Wildebeest was the most common species in terms of the proportion of bushmeat packages recorded (Figure 4.6).

Figure 4. 6: Wildlife species contribution to total bushmeat packages recorded in the dry and the wet season.

More bushmeat packages were recorded during the dry season (77.2%) than in the wet season (22.8%). The majority of the packages (76.5%) were dried meat, and the rest were fresh meat (23.5%). Significantly more hunters claimed to hunt mostly inside the SNP (65%) followed by WMA (22.5%) and Game Reserves (12.5%) (Kruskal-Wallis test; H = 44.507; P < 0.001).

Hunters (45.83%, n = 96) and bushmeat traders (54.17%, n = 96) had different roles in the whole process of hunting and trading bushmeat although most hunters (61.36%, n = 44) were also bushmeat traders. Hunters were responsible for killing the animal, do the initial processing (removing the skin and cutting the meat into portions), transporting bushmeat from the bush to the village and may upgrade the meat by drying it (Plate 4.1) to facilitate long distance transportation.

67

Plate 4. 1: Drying bushmeat packages outside hunter’s household in Robanda village.

Plate 4. 2: Bushmeat package exposed for sun drying outside a regular household in Robanda village.

68

Plate 4. 3: Bushmeat package (Impala) measured from hunter’s household in Robanda village.

Plate 4. 4: Bushmeat package (Thomson’s gazelle) sold in one of the regular household at Robanda village.

69 Bushmeat was either consumed (47%) in the household or sold (53%) within the village and outside the village. Bushmeat business was conducted door to door within the village also in neighbouring villages and in big towns as shown in Figure 4.7.

Figure 4. 7: Map the study area and its location in Tanzania (left) showing the study villages (Robanda, Rwamkoma and Kowak) and the bushmeat market centers indicated with red dots.

Bushmeat prices (Table 4.5) were different following the type of wildlife species, state of the meat (dried or fresh), location (distance from protected area boundary) and weight or size of the package. Hunters were selling the meat to bushmeat traders as whole sellers and also to consumers or regular households within the village as retail sellers. Bushmeat traders also upgrade the meat (cut into smaller pieces or packages) depending on the needs of the customers, transporting it and sell to end consumers. Transportation of bushmeat to villages occurred mostly during the night and was done mainly by foot (39.58%), followed by motorcycle (31.25%) and bicycle (29.17%).

70 Table 4. 5: Average bushmeat price per 1kg of different wildlife species (1 US$ = TZS 2,250) Species name Mean price US$ TZS Giraffe (Giraffa camelopardalis) 4.00 9,000 Impala (Aepyceros melampus) 1.77 3,977 Grant‟s gazelle (Nanger granti) 1.69 3,800 Thomson‟s gazelle (Eudorcas thomsonii) 1.58 3,550 Topi (Damaliscus lunatus) 1.53 3,438 Common eland (Taurotragus oryx) 1.47 3,300 Blue wildebeest (Connochaetes taurinus) 1.40 3,139 Plains zebra (Equus quagga) 1.04 2,330 Source: Field data (2017-2018)

4.8.2 Contribution of other income sources to total household income as a function of distance from PA boundary Overall, average total household income was significantly higher in Robanda than in Rwamkoma and Kowak (Figure 4.8; Kruskal-Wallis test; H = 10.975, P = 0.004). A post-hoc Dunn‟s test reveals significant differences between Robanda and Rwamkoma (Dunn‟s tests; Z = -2.236; P = 0.051) as well as between Robanda and Kowak (Z = 3.04; P = 0.007), while the difference between Rwamkoma and Kowak was not significant (Z = 0.68; P = 0.496).

Overall, regular households on average obtained significantly higher total household income than hunter and bushmeat trader households (Wilcoxon test; W = 5741; P = 0.0074). However, comparisons at the village level revealed no significant differences Table 4.6.

71

Figure 4. 8: Average total income in hunter and bushmeat trader and regular households at increasing distance from the PA boundary.

Table 4. 6: Test of significance differences in mean income and reliance on income from different sources for hunter and bushmeat trader and regular households, overall and in each village. Overall Robanda Rwamkoma Kowak Income Source W P W P W P W P Crop income 8409 0.018 2175 0.074 186 0.11 199.5 0.43 Crop reliance 8733 0.003 2211 0.047 296 0.62 170.5 0.21 Livestock 8818.5 0.001 2550 <0.001 211 0.26 258 0.55 income Livestock 8835.5 <0.001 2577 <0.001 265 0.94 259 0.54 reliance Business and 7933.5 0.169 2648 <0.001 196.5 0.02 227 0.89 wages income Business and 7975.5 0.146 2886 <0.001 201.5 0.03 304.5 0.13 wages reliance Total household 5741 0.007 1761.5 0.613 174 0.07 237 0.96 income (W = Wilcoxon test and P = P value)

72 Other income sources (agricultural and non-agricultural income) and their contribution to hunters and traders‟ total household income also differed between study villages (Figure 4.4 and Figure 4.5). Crop income was significantly lower in the closest village Robanda than in the intermediate village, Rwamkoma and distant village, Kowak (Kruskal-Wallis test; H = 29.43; P < 0.001). A post-hoc Dunn‟s test reveals significant differences between Robanda and Rwamkoma (Z = 5.214; P < 0.001) as well as between Robanda and Kowak (Z = -2.091; P = 0.037), while the difference between Rwamkoma and Kowak was non-significant (Z = 2.103; P = 0.071). Household reliance on crop income was significantly lower in the closest village Robanda than in the intermediate village, Rwamkoma and distant village, Kowak (Kruskal-Wallis test; H = 30.508; P < 0.001). The differences between Robanda and Rwamkoma (Dunn‟s tests; Z = 5.15; P < 0.001) as well as between Robanda and Kowak (Z = -2.58; P = 0.02) were significant, while the difference between Rwamkoma and Kowak was insignificant (Z = 1.67; P = 0.094).

Overall, regular households income from crop production was on average significantly higher than hunter and bushmeat trader households (68.5 ± 6.01 vs 37.4 ± 3.77) (Wilcoxon test; W = 8409; P = 0.018) whereas, comparisons at the village level revealed no significant differences Table 4.6. Overall, household reliance on crop income was significantly higher in regular households (Figure 4.10) than in hunter and trader households (Figure 4.5) (Wilcoxon test; W = 8733; P = 0.003). At the village level regular households also had a significantly higher reliance on crop income than hunters and traders in the closest village, Robanda (W = 2211; P = 0.047) but not in the intermediate village, Rwamkoma (W = 296; P = 0.615) and distant village, Kowak (W = 170.5; P = 0.21).

Hunters and bushmeat traders‟ income from livestock production also differed significantly between villages (Figure 4.4; Kruskal-Wallis test; H = 16.165; P < 0.001). The differences were significant between Robanda and Rwamkoma (Dunn‟s tests; Z = 3.918; P < 0.001) as well as between Rwamkoma and Kowak (Z = 3.16; P = 0.003), while the difference between Robanda and Kowak was not significant (Z = 0.445; P = 0.657). Reliance on livestock income was significantly higher in the intermediate village, Rwamkoma than in the closest village, Robanda and the distant

73 village, Kowak (Figure 4.5; Kruskal-Wallis test; H = 17.553; P < 0.001). A post-hoc Dunn‟s test reveals significant differences between Robanda and Rwamkoma (Dunn‟s tests; Z = 4.09; P < 0.001) as well as between Rwamkoma and Kowak (Z = 3.26; P = 0.002), while the difference between Robanda and Kowak was not significant (Z = 0.42; P = 0.68).

Figure 4. 9: Average crop, livestock and wage and business income of regular households at increasing distance from the PA boundary.

Overall, hunters and bushmeat traders had significantly lower income from livestock production than regular households (164 ± 18.1 vs 53.8 ± 6.36) (Figure 4.4 and 4.9; Wilcoxon test; W = 8818.5; P = 0.001). Similar significant differences were observed in the closest village, Robanda (Wilcoxon test; W = 2550.5; P < 0.001) while no significant difference was observed in the intermediate village, Rwamkoma (W = 211; P = 0.257) and in the distant village, Kowak (W = 258; P = 0.553). Overall, household reliance on livestock income was significantly higher in regular households (Figure 4.10) than in hunter and trader households (Figure 4.5) (Wilcoxon test; W = 8835.5; P < 0.001). At the village level regular households also had significantly higher livestock reliance than hunters and traders in the closest

74 village, Robanda (W = 2577; P < 0.001) but not in the intermediate village, Rwamkoma (W = 265; P = 0.938) and in the distant village, Kowak (W = 259; P = 0.537) Table 4.6.

Figure 4. 10: Reliance (percentage contribution to total household income) of livestock, crop and wage and business income of regular households at increasing distance from the PA boundary.

Hunter and bushmeat trader households income from business, wage work and other sources (pension and remittances) also differed between study villages (Figure 4.4) with significantly lower income in Rwamkoma than Robanda and Kowak (Kruskal- Wallis test; H = 8.271; P = 0.016). The differences between Robanda and Rwamkoma (Dunn‟s test; Z = -2.579; P = 0.019) as well as between Rwamkoma and Kowak (Z = -2.608; P = 0.027) were significant, while the difference between Robanda and Kowak was insignificant (Z = -0.965; P = 0.335). Reliance on business, wage and other income was also significantly lower in the intermediate village, Rwamkoma than in the distant village, Kowak and in the closest village, Robanda for hunters and bushmeat traders (Figure 4.5; Kruskal-Wallis test; H = 11.297; P = 0.004). A post-hoc Dunn‟s test reveals significant differences between Robanda and

75 Rwamkoma (Z = -2.58; P = 0.02) as well as between Rwamkoma and Kowak (Z = - 3.3; P = 0.003), while the difference between Robanda and Kowak was not significant (Z = -1.85; P = 0.065).

Overall, income from business, wage and other sources and reliance on this income did not differ significantly between hunter and trader and regular households (Table 4.6: business and wage income; Wilcoxon tests; W = 7933.5; P = 0.169 and business and wage reliance; W = 7975.5; P = 0.146). Comparisons at the village level showed that regular households obtained significantly higher income from business, wage and other sources than hunters and bushmeat traders in the closest village, Robanda (Wilcoxon test; W = 2648; P < 0.001) and in the intermediate village, Rwamkoma (W = 196.5; P = 0.019) but not in the distant village, Kowak (W = 227; P = 0.894) (Figure 4.4 and 4.9). Similar significant differences were also observed for reliance on business, wage and other income in Robanda (Wilcoxon test; W = 2886; P < 0.001) and Rwamkoma (W = 201.5; P = 0.029) but not in Kowak (W = 304.5; P = 0.132) (Figure 4.5 and 4.10).

4.9 Factors influencing household participation in hunting and trading bushmeat and household bushmeat income reliance The results of the Heckman sample selection model evaluating factors associated with household participation in hunting and trading bushmeat and predicting these households reliance on bushmeat income are presented in Table 4.7. The selection part of the model revealed that the probability of households participation in hunting and the bushmeat trade is associated with distance from the village to the PA boundary and the gender of the household head. Female-headed households are less likely to participate in hunting and trading bushmeat and the further the village is from the PA boundary, the less likely that households participate in hunting and trading bushmeat. Other factors such as household cash income, household size and age of the household head were not significantly associated with participation in hunting and trading bushmeat.

76 Factors influencing household reliance on bushmeat income were evaluated in the outcome part of the model. The outcome part of the model revealed that age and gender of the household head and distance to PA boundary were negatively associated with bushmeat income reliance, whereas none of the other predictors was significantly associated with bushmeat income reliance.

Table 4. 7: Heckman sample selection model predicting household participation in hunting and trading bushmeat and reliance on bushmeat income. Variables Coefficients SE T P Selection equation Distance to PA boundary (km) -0.0274 0.0046 -6.001 <0.001 Gender of the HH head -1.0280 0.1883 -5.461 <0.001 HH cash income (US$ per capita) 0.00004 0.0003 0.123 0.902 Age of HH head -0.0458 0.0546 -0.840 0.402 Number of household members -0.0322 0.0428 -0.752 0.453 Intercept 2.0710 0.3988 5.192 <0.001 Outcome equation Distance from PA boundary (km) -0.0335 0.0122 -2.754 0.006 Gender of the HH head -1.0550 0.4469 -2.360 0.019 Age of HH head -0.1235 0.0549 -2.252 0.025 HH cash income (US$ per capita) 0.00009 0.0005 0.207 0.837 Actor group (hunters or traders) 0.0957 0.1214 0.788 0.431 Intercept 1.5430 0.4218 3.657 <0.001 InvMillsRatio 1.3366 0.5844 2.287 0.023 Rho 1.2289

Multiple R-Squared: 0.229, Adjusted R-Squared 0.177; N = 246

77 CHAPTER FIVE

DISCUSSION

5.1 Overview This chapter discusses the study findings and give interpretations. The discussion is organised around the research hypotheses. It discusses and interprets information obtained by this study and also integrates the discussion with other similar studies elsewhere. It explains on the variation in bushmeat consumption frequencies between seasons and across villages located along a gradient of distance from the SNP boundary. The factors influencing household dependence on bushmeat consumption and income are also discussed in this chapter.

5.2 Hypothesis 1: Bushmeat consumption frequency is higher during the dry season than the wet season The study assumption that bushmeat was more consumed during the dry season was supported by information provided by both schoolchildren and adult respondents. During the dry season the availability of bushmeat increases due to migration of herbivores including wildebeests and zebras which migrate seasonally in the area as a response to resources availability (Holmern, Mkama, Muya, & Røskaft, 2006; Holmern, Muya, et al., 2007). Animals have to migrate in respose to seasonal variation which influences the availability of resources (Sinclair et al., 2015).

Such movements expose the animals into many different areas both inside and outside protected areas and this gives more chance for illegal hunting hence increase the availability of bushmeat to local communities living around (Rentsch & Packer, 2015). It has been also documented by previous studies that, more incidences of illegal hunting has been reported to occur during the dry season (Loibooki et al., 2002; Nyahongo et al., 2009). In this study it was found that, the increased illegal hunting during the dry season is due to increased openness and right weather conditions and also lack of farming activities which occurs mostly during the wet season. Other reasons include higher abundance and migratory behaviour of herbivores such as wildebeest (Holmern et al., 2006) which expose them more and therefore increases the risks of being hunted for bushmeat (Sinclair et al., 2015).

78 In the wet season bushmeat consumption was relatively low probably because of little migration which reduces the availability of herbivores to hunters. During the wet season, the resident herbivores were the primary source of bushmeat (Rentsch et al., 2015). At this time, the consumption of domestic meat was increased especially in the closest village. However, the consumption of other meat types particularly fish was also low in the wet season which may be attributed to poor infrastructure and roads being inaccessible during heavy rain hindering the transportation of fish, including sardines (Nyahongo et al., 2009).

5.3 Hypothesis 2: Bushmeat consumption frequency decreases with increasing distance from protected area boundary Bushmeat consumption frequency decreases as the distance between village and park boundary increases. The study found that, households in the closest village consume bushmeat at the average of four times per week and two times per week for both dry and wet seasons respectively. However, in the intermediate and distant village bushmeat consumption was very rare (Figure 4.1). Similar trend has also been observed in previous studies (Fischer et al., 2014; Nyahongo et al., 2009). In western Serengeti households depend on bushmeat as a source of meat protein food (Barnett, 1997) and about 45-60% of the households consume bushmeat regularly (Rentsch et al., 2015). This show how households living close to wildlife areas relies on wildlife resources which also help to balance the costs incurred due to problem animals (Bitanyi et al., 2012; Galvin et al., 2008). Morover, higher consumption frequency in the closest village is associated with animal abundance (Holmern, Røskaft, Mbaruka, Mkama, & Muya, 2002) and also cultural influence (Knapp et al., 2017). Bushmeat consumption has cultural values (Kideghesho, 2008) which drive their consumption rates for example the Ikoma people in Serengeti district (Mfunda & Røskaft, 2010). The distant villages obtain bushmeat from hunters through the bushmeat trade (Mwakatobe et al., 2012) due to limited hunting opportunities associated with long distance from wildlife areas. However, the recent increased focus on law enforcement has increased the likelihood of being fined for trading bushmeat and also increased the penalty. This has led to increased challenges in transportation of bushmeat from villages close to protected areas to villages located further. Perhaps, as a result, the frequency of bushmeat

79 consumption in the intermediate village, Rwamkoma and in the distant village, Kowak was found to be low. Hence, the results support Hypothesis 2 that bushmeat is frequently consumed in villages close to wildlife areas.

5.4 Hypothesis 3: Schoolchildren report higher bushmeat consumption frequency than adults A comparison between respondents on reported bushmeat consumption frequencies show that children reported more consumption than adults. Although the assumption was that adult might report low consumption due to sensitivity of the information, ability to recall was also a reason for some adults to report lower consumption frequency. However, the differences in their data collection methods can also influence the results obtained. Adults were asked to recall food consumption on weekly bases while children were asked on daily bases except for weekend. Dietary recall survey was used to collect information from children but for adults it was limited by the nature of the respondents although it is among the most recommended method (Brashares et al., 2011; Knapp et al., 2010; Rentsch & Packer, 2015). Similar trend has also been observed by previous studies that, adults behave strategically when responding to sensitive issue such as illegal bushmeat hunting (Rentsch et al., 2015). However, there is no logical reason for children to exaggerate their consumption frequencies. Therefore, the study assumption was supported by the findings.

5.5 Hypothesis 4: Bushmeat consumption frequency is associated with household socioeconomic and other characteristics The model predicted the factors influencing bushmeat consumption in households as the study hypothesised. Among socioeconomic factors, age of the respondent was found to have significant influence on bushmeat consumption in households. Other significant factors were season of the year, distance between village and park boundary and consumption of other meat sources. Sardines (small sun-dried fish) were mostly consumed as an alternative to bushmeat particularly in the intermediate and distant villages for both seasons. It was less expensive and therefore available for most people compared to other fish types and domestic meat since majority are

80 economically poor (Kiffner et al., 2015; Ndibalema & Songorwa, 2007; Nyahongo et al., 2009). The consumption of domestic meat was relatively low in almost all villages due to its higher prices compared to other meat sources. Furthermore, in the closest village Robanda, there were no sign of butchers selling beef but only rather few groceries selling goat‟s meat with beers, which usually is considered a luxury good. However, even the claimed goat‟s meat sold in Robanda cannot be certainly sure because they also sell medium sized herbivores such as gazelles and impalas. In the intermediate and distant villages, different types of animal meat were sold regularly in local butchers located within the village and also in the village market during auction.

5.6 Hypothesis 5: Bushmeat hunting and household reliance on bushmeat income decreases with distance from PA boundary The study findings support hypothesis 5 in both statistics through comparison of means and model evaluation. Household involvement in bushmeat hunting activities and reliance on bushmeat income was inversely related to distance. Hunting was illegal and highly performed by local people from the closest village, Robanda. Households found in far distance were less engaged in hunting and trading bushmeat probably due to higher transportation cost and the elevated risks associated with the activity (Knapp, 2012).

Household dependency on bushmeat income was higher in Robanda, the closest village than in other villages located far distance to the PA boundary. The higher abundance of wild animals influences people to do illegal hunting in the closest village since hunters can get access to the animals easily during their migration (Nyahongo et al., 2009; Rentsch & Packer, 2015). The hunting was found to be common in western Serengeti as most people depend on bushmeat as their source of meat protein and also income for the household (Mfunda & Røskaft, 2010).

In distant villages, household dependency on bushmeat income was significantly lower than in the closest village (Figure 4.1) for various reasons. Most households in the intermediate and in the distant villages were less involved in bushmeat hunting activities due location and also low experience in such activities. In such villages,

81 reliance on bushmeat income was relatively lower compared to other income sources (Figure 4.5). This can be associated with the costs of doing illegal hunting including fines and imprisonment (Knapp et al., 2017), all these influences peoples‟ decision to participate in such activities based on cost and benefit analysis. The costs are higher for hunters from villages located further from PAs, which drives the decision for selecting other livelihood strategies rather than illegal hunting.

However, information from previous studies revelead that poverty was among the main factors influencing people to engage in illegal bushmeat hunting activities (Kideghesho, 2009). This study found that reliance on bushmeat income was higher in the closest village and the contribution of other income sources was lower while in the villages located far distance to PA boundary the contribution of other income sources were significantly higher than bushmeat income (Figure 4.5). However, household income portfolios were different in all the three villages because of various reasons including differences between geographical locations and background of the local people.

In the closest village Robanda, there were more economic opportunities than in other study villages due to location (i.e being close to PAs) where tourists and other people can visit the village more frequently for different purposes including tourism and conservation activities (Kyando et al., 2019). However, the increased economic opportunities have led to laziness in some people as a result some households are very poor while other few are very rich. In the intermediate village Rwamkoma, agricultural production was the main economic activity for most households. The area has good fertile soil and relatively reliable rainfall patterns that allows cultivation at large scale but the majority do small-scale farming (Estes et al., 2012) due to inadequate skills and capital. The distant village Kowak depends on both agricultural production and small-scale business because the area does not have good fertile soil and reliable rainfall. Therefore, it requires households to rely on more than one income sources in order to survive in the area.

82 5.6.1 Source of bushmeat The study found that, wildebeest was the main source of bushmeat consumed by households in western Serengeti as it constituted higher propotion of bushmeat packages observed (Figure 4.6). A study by Rentsch and Packer (2015) also found similar trend and reported that two third of the animals hunted for bushmeat was wildebeest. Hunting occurs mostly during the dry season when groups of herbivores particularly wildebeests and zebras migrate. The migration greatly influences the availability of bushmeat for household consumption in the area (Nyahongo et al., 2009). The animals migrate without boundaries and therefore cross within and outside protected areas.This increases the probability of being hunted as they also pass in areas occupied by the local people (Sinclair et al., 2015). The closest village, Robanda act as a migratory route for these migratory herbivores which normally pass through the area every year (Mwakatobe et al., 2012; Rentsch & Packer, 2015). Hunters use this opportunity especially those who live close to the PA boundary. This gives some explanation for more bushmeat packages (80.54%) found in the closest village, Robanda compared to other villages located further.

More hunting (65%) occurred in Serengeti National Park (SNP) compared to the Wildlife Management Area (22.5%) and Game Reserves (12.5%) due to the large area of the SNP where hunters can hide in remote places escaping the rangers. Also the communities living very close to the park boundary were found to rely more on SNP as it was easy to reach the area. Snaring was the most hunting method used in addition to others including the use of motorcycles, pitfall traps, dogs and torches. Previous studies also observed that hunters prefer snaring than other hunting methods because of its availability, low cost and easy to use (Holmern et al., 2006; Knapp, 2012; Nyahongo et al., 2006). However, the use of wire snares in hunting is a threat to wild animals because it can kill animals regardless of age, sex and type (Gideon, 2014). It can cause harm to other animals such as carnivores and also young animals which are not targeted by illegal hunters when they pass through the trap (Loibooki et al., 2002).

83 Bushmeat was mainly sold as dried meat (76.5%) and few (23.5%) as fresh meat. Most of the bushmeat sold was sun dried in order to simplify handling when sold door to door in households and also to avoid spoilage. Bushmeat business was conducted secretly and there was no formal bushmeat market in the area. The secrete markets were also outside the study villages including big towns such as Mugumu, Tarime, Nyamswa, Bunda, Utegi, Shirati and Kenya in some towns located close to the bourder between Tanzania and Kenya (Figure 4.7).

5.7 Hypothesis 6: Socioeconomic factors influencing household participation in hunting and reliance on bushmeat income Gender was among the significant factors influencing people to participate in bushmeat hunting activities. Females were less involved in bushmeat hunting activities due limited labour surplus and low experience in such activities because they are commonly performed by men. Females were mostly involved in other economic activities and they relied on income from agriculture, small scale business and employment. However, other factors such as age of the household head, household size and household income were not significant.

Economically, the majority of people were poor both hunters/bushmeat traders and regular households (Table 4.4 and Figure 4.8) and this was the reason for household income to be insignificant among predicting factors. The reasons for this economic hardship in most people include limited livelihood opportunities, inadequate skills because of low education as most people had primary education (Table 4.2 and 4.4) and constrains on agricultural activities such as poor farming practices and unpredictable weather conditions.

Household dependency on bushmeat income was negatively related to the age, which means that younger household heads were less dependency on this income compared to older men. Older males had lower education which inhibits their ability to do other activities for their household income and therefore relied on bushmeat since they have experience in hunting. However, other studies have found a negative association between household income and bushmeat income reliance meaning that bushmeat is more important for poor households (Nielsen et al., 2018).

84 In this study such relationship was not observed statistically but still bushmeat reliance is negatively associate with wealth (Brashares et al., 2011) and people who rely on bushmeat have limited sources of income (FAO, 2015; Fischer et al., 2014). This study found that, people were economically poor and therefore engaged in bushmeat hunting activities due to limited income sources to sustain their livelihoods. Hunters and bushmeat traders had significantly lower income from other sources compared to bushmeat income (Figure 4.4). Previous studies have also reported some people are engaged in bushmeat hunting activities because of the need for cash income (Knapp, 2007, 2012).

In the closest village, bushmeat income was higher than other income sources in hunter and bushmeat trader households and their income was significantly lower compared to that of regular households. Regular households had more income from agricultural production (crop and livestock income) and non-agricultural activities (business, wage and other income). However, in this village the agricultural activities were highly affected by the problem animals through crop raiding and livestock depredation (Galvin et al., 2008; Kyando et al., 2019; Mwakatobe, 2013).

Despite the costs incurred by the local people caused by problem animals, being close to PA boundary is advantageous for the village as it generate income from tourism activities and also a source of employment to local people (Kyando et al., 2019). This village also manages a WMA where income generated goes back to the village for community development projects. Due to such huge income from the WMA, Robanda village is considered among the richest villages in the district. However, individual households were still poor despite the developments attained by the village since communal projectes deliver communal benefits which means the benefits are indirect to people. The need for direct individual benefit from conservation was among the factors associated with illegal bushmeat hunting in Robanda despite the benefits derived from conservation particularly from Ikona WMA.

85 Farming was the main economic activity for most people in the area despite the challenges associated with inadequate skills and poverty which cause people to practise small scale farming. Also the problem of crop raiding and low output discourage people to rely on agriculture and therefore opt for other activities including bushmeat hunting. Improving the agricultural sector can help to encourage people to stick on such activities and therefore reduce bushmeat hunting activities. A study by Rentsch and Packer (2015) reported that people with more labour intensive economic activities had less time to invest in illegal bushmeat hunting activities. This may be part of the explanation for people in the intermediate village, Rwamkoma to rely on other income sources rather than bushmeat.

In addition to economic factors, other factors are also influencing people to rely on bushmeat for example cultural factors (Kideghesho, 2008). Some society believes that bushmeat is healthier than domestic meat and it test good (Kideghesho, 2008, 2009). Furthemore, bushmeat is cheaper and therefore serve as alternative meat for domestic meat types which are expensive, particularly in areas close to wildlife areas (Fischer et al., 2014; Manyama, Nyahongo, et al., 2019).

86 CHAPTER SIX

CONCLUSION AND RECOMMENDATIONS

6.1 Conclusion The results from this study reveal that bushmeat is an important source of meat protein in households living close to park boundary in western Serengeti and it is more consumed during the dry season. The consumption of bushmeat was significantly higher in the closest village and the consumption decrease as distance between villages and park boundary increase. A comparison between respondents show that children reported more bushmeat consumption than adults in all the three villages. Children are less sensitive to illegal bushmeat hunting activities since they are still young to participate in such activities. This provides a free chance for them to give more reliable information about bushmeat consumption in their households. The consumption frequency of other meat types particularly fish was significantly higher in the intermediate and distant villages than in Robanda, the village closest to SNP. Fish can also be used as an alternative meat source in the closest village where bushmeat was highly consumed.

Bushmeat is also an important source of income for hunters and bushmeat traders, particularly in the village close to the protected area boundary. The participation in bushmeat hunting activities is negatively associated with distance to the PA boundary. Other factors such as age and gender of the household head were also significant in determine the likelihood of being a hunter or bushmeat trader. Female- headed households and younger household heads were less involved in bushmeat activity and therefore relied less on bushmeat income. This suggests that the occurrence of illegal hunting activities may be reduced over time if the young men keep on relying on other income sources. Hence, efforts to reduce incidence of such activities should target older males close to the PA boundary. Alternative sources of meat and income will also help to reduce household dependence on bushmeat in western Serengeti.

87 6.2 Recommendations This study provides valuable insights for targeting policies to reduce illegal hunting and bushmeat trade through promoting alternative meat and income sources. The study recommends on increasing production and availability of other meat sources in order to reduce reliance on bushmeat. The governemt should support the local communities in establishing fish farming and poultry projects in order to promote production of alternative meat sources and to increase availability for local consumption. Through establishment of fish farms more people can get access to fish as it occurs in other areas (Brashares et al., 2004) and therefore serve as an alternative to bushmeat. It will also help to enhance sustainable fishing in natural rivers and lakes particularly Lake Victoria which is the main source of fish in the area. The study found that sardines were among the alternative meat sources that most people relie on particularly in the intermediate and distant villages. Its availability can be increased even to villages located close to park boundary so that it can be used as an alternative meat to bushmeat in such areas. However, concerns about sustainable utilization is of great importance in order to reduce impacts on fish stocks (Rentsch & Damon, 2013).

Until then increasing the number of alternative livelihood opportunities for older male-headed households in villages close to the PA boundary may reduce reliance on bushmeat income and hence the combined hunting pressure on wildlife populations in the GSE. Increasing the availability and reducing the price of alternative meat types can also help to reduce bushmeat demand and consumption in the area (Walelign, Nielsen, & Jakobsen, 2019). In addition to law enforcement, which is currently the standard response and considered the most effective approach to control illegal hunting by some scholars (Rentsch & Damon, 2013), other strategies may need to be used to control illegal hunting in western Serengeti.

This will add efforts in reducing illegal behaviours in areas were the traditional approach of protected area management through fences and fines has failed (Barrett & Arcese, 1995). The strategies should focus on key stakeholders engaged in hunting and trading bushmeat as identified in this study to cut the supply of bushmeat and reduce the bushmeat trade. Incentive schemes should address the root causes of

88 people engaging in illegal hunting (Duffy et al., 2019). Such strategies could involve providing capital for small-scale business, employment opportunities in the tourism sector and conservation jobs targeted for hunters and bushmeat traders and promoting the consumption of other meat types.

The agricultural sector should be improved through the use of advanced farming practices and livestock keeping so that people can cultivate more profitable crops and keep few livestock which are more beneficial. Diversification of economic activities can also help to increase economic opportunities for local people. Some might require short training particularly technical works such as welding, lumberling etc. Short training in the form of adult education can help to promote diversification of economic activities in order to improve peoples‟ knowledge which can help in selecting the best livelihood strategies for their wellbeing. The human wildlife conflicts should be reduced and wildlife related costs to local people should be compensated. Conservation benefits should be increased especially to households living close to protected areas as a strategy to reduce negative attitudes towards wildlife conservation and illegal behaviours. Wildlife Managed Areas (WMAs) should provide adequate benefits to villagers and seek to promote community development and improved livelihoods together with natural resources conservation. All these can help to increase conservation benefits to local people and therefore, reduce household reliance on bushmeat income and illegal behaviours.

6.2.1 Recommendations for further research i) Research on wildlife species hunted for bushmeat Wildlife species that are illegally hunted for bushmeat should be studied in order to assess the effects of poaching to their populations, behaviours and habitats. Some of them have been identified in this study (Figure 4.6) but their populations, behaviours and habitats were not studied for assessment of negative impacts associated with illegal hunting. Also the impacts to other species and ecosystem in general should be studied since poaching activities involves destractive methods such as pitfall traps, snaring, the use of motorcycle to chase animals and others.

89 ii) Assessment of altenative meat sources Analysis of alternative meat sources should be conducted in villages bordering protected areas in order find suitable alternatives that will be feasible and acceptable by the local people. Fish particularly sardines was among the alternative source identified under this study however, it needs further analysis before being implemented.

iii) Research on how to manipulate bushmeat market Potential areas for bushmeat market should be identified and controlled both in villages and in big towns. Also the price of other meat types should be fearly reduced so that it can help to manipulate bushmeat demand and hence reduce illegal hunting.

iv) Diversification of economic activities of local people in western Serengeti Studies on how to diversify economic activities in villages‟ adjacent protected areas can help to increase alternative income sources for hunter and bushmeat trader households and therefore reduce dependence on bushmeat income.

90 REFERENCES

Acheson, M. J. (2002). Rational choice, culture change, and fisheries management in the gulf of Maine Research in Economic Anthropology (Vol. 21, pp. 133- 159): Emerald Group Publishing Limited.

Adefalu, L. L., Oladipo, F. O., Usman, B. A., Babalola, F. D., Amusa, T. O., & Egere, S. (2012). Rural women perception on bushmeat trade around Kainji Lake national park, Niger state, Nigeria. Ethiopian Journal of Environmental Studies and Management, 5(1), 20-25.

Ahmadi, S., Maman, S., Zoumenou, R., Massougbodji, A., Cot, M., Glorennec, P., & Bodeau-Livinec, F. (2018). Hunting, Sale, and Consumption of Bushmeat Killed by Lead-Based Ammunition in Benin. International journal of environmental research and public health, 15(6), 1140-1152.

Andimile, M., Tim, C., & Mulder, M. B. (2012). Bushmeat consumption in western Tanzania: A comparative analysis from the same ecosystem. Tropical Conservation Science, 5(3), 352-364.

Arcese, P., Hando, J., & Campbell, K. (1995). Historical and Present-day Anti- Poaching Efforts in Serengeti. In A. R. E. Sinclair & P. Arcese (Eds.), Serengeti II: Dynamics, Management, and Conservation of an Ecosystem (pp. 506-533). Chicago and London: University of Chicago Press.

Bakkegaard, K. R., Nielsen, R. M., & Thorsen, J. B. (2016). Household determinants of bushmeat and eru (Gnetum africanum) harvesting for cash in the Democratic Republic of Congo. Environment, Development and Sustainability, 19, 1-19.

Baranowski, T., Islam, N., Baranowski, J., Martin, S., Beltran, A., Dadabhoy, H., . . . Subar, A. F. (2012). Comparison of a Web-Based versus Traditional Diet Recall among Children. Journal of the Academy of Nutrition and Dietetics, 112(4), 527-532.

91 Barnett, R. (1997). Food for Thought: The Utilization of Wild Meat in Eastern and Southern Africa. Retrieved from Nairobi, Kenya:

Barrett, B. C., & Arcese, P. (1995). Are Integrated Conservation-Development Projects (ICDPs) Sustainable? On the conservation of large mammals in sub- Saharan Africa. World Development, 23(7), 1073-1084.

Barrett, B. C., & Arcese, P. (1998). Wildlife Harvest in Integrated Conservation and Development Projects: Linking Harvest to Household Demand, Agricultural Production, and Environmental Shocks in the Serengeti. Land Economics, 74(4), 449-465.

Bitanyi, S., Nesje, M., Kusiluka, L. J., Chenyambuga, S. W., & Kaltenborn, B. P. (2012). Awareness and perceptions of local people about wildlife hunting in western Serengeti communities. Tropical Conservation Science, 5(2), 208- 224.

Brashares, S. J., Arcese, P., Sam, M. K., Coppolillo, P. B., Sinclair, A. R., & Balmford, A. (2004). Bushmeat hunting, wildlife declines, and fish supply in West Africa. Science, 306(5699), 1180-1183.

Brashares, S. J., Golden, D. C., Weinbaum, Z. K., Barrett, B. C., & Okello, V. G. (2011). Economic and geographic drivers of wildlife consumption in rural Africa. PNAS, 108(34), 13931–13936.

Brooks, S. E., Kebede, B., Allison, E. H., & Reynolds, J. D. (2010). The Balance of Power in Rural Marketing Networks: A Case Study of Snake Trading in Cambodia. 46(6), 1003 - 1025.

Campbell, K., & Hofer, H. (1995). People and Wildlife: Spatial Dynamics and Zones of Interaction. In A. R. E. Sinclair & P. Arcese (Eds.), Serengeti II: Dynamics, Management, and Conservation of an Ecosystem (pp. 534-570): University of Chicago Press.

CBD. (2011). Livelihood Alternatives for the Unsustainable use of Bushmeat. Retrieved from Montreal:

92 Ceppi, L. S., & Nielsen, R. M. (2014). A comparative study on bushmeat consumption patterns in ten tribes in Tanzania. Tropical Conservation Science, 7(2), 272-287.

Certo, S. T., Busenbark, R. J., Woo, H.-s., & Semadeni, M. (2016). Sample selection bias and Heckman models in strategic management research. Strategic Management Journal, 37, 2639–2657.

Chaves, W. A., Wilkie, D. S., Monroe, M. C., & Sieving, K. E. (2017). Market access and wild meat consumption in the central Amazon, Brazil. Biological Conservation, 212(Part A), 240-248.

Coad, L., Abernethy, K., Balmford, A., Monica, A., Airey, L., & Milner-Gulland, E. J. (2010). Distribution and use of income from bushmeat in a rural village, central Gabon. Conservation Biology, 24(6), 1-9.

Costello, C., Burger, N., Galvin, A. K., Hilborn, R., & Polasky, S. (2008). Dynamic Consequences of Human Behaviour in the Serengeti Ecosystem. In A. R. E. Sinclair, C. Packer, A. R. S. Mduma, & M. J. Fryxell (Eds.), Serengeti III: Human Impacts on Ecosystem Dynamics Chicago and London: The University of Chicago Press.

Cowlishaw, G., Mendelson, S., & Rowcliffe , M. J. (2005). Structure and Operation of a Bushmeat Commodity Chain in Southwestern Ghana. Conservation Biology, 19(1), 139 - 149.

Crookes, D. J., & Milner-Gulland, E. J. (2006). Wildlife and economic policies affecting the bushmeat trade: a framework for analysis. South African Journal of Wildlife Research, 36(2), 159–165.

Damania, R., Milner-Gulland, E. J., & Crookes, D. J. (2005). A bioeconomic analysis of bushmeat hunting. Proceedings. Biological sciences, 272(1560), 259-266.

93 Dawson, C. (2007). A Practical Guide to Research Methods: A user-friendly manual for mastering research techniques and projects (3 ed.). Oxfordshire: How to Books Ltd. de Merode, E., & Cowlishaw, G. (2006). Species Protection, the Changing Informal Economy, and the Politics of Access to the Bushmeat Trade in the Democratic Republic of Congo. Conservation Biology, 20(4), 1262–1271. de Merode, E., Homewood, K., & Cowlishaw, G. (2004). The value of bushmeat and other wild foods to rural households in extreme poverty in Democratic Republic Congo. Biological Conservation, 118, 573-581.

Diep, C. S., Hingle, M., Chen, T.-A., Dadabhoy, H. R., Beltran, A., Baranowski, J., . . . Baranowski, T. (2015). The Automated Self-Administered 24-Hour Dietary Recall for Children, 2012 Version, for Youth Aged 9 to 11 Years: A Validation Study. Journal of the Academy of Nutrition and Dietetics, 115(10), 1591-1598.

Dudley, N. (2008). Guidelines for applying protected area management categories. Retrieved from Gland, Switzerland:

Duffy, R., Massé, F., Smidt, E., Marijnen, E., Büscher, B., Verweijen, J., . . . Lunstrum, E. (2019). Why we must question the militarisation of conservation. Biological Conservation, 232, 66-73.

Edderai, D., & Dame, M. (2006). Short Communication: A census of the commercial bushmeat market in Yaounde, Cameroon. Oryx, 40(4).

Estes, A. B., Kuemmerle, T., Kushnir, H., Radeloff, V. C., & Shugart, H. H. (2012). Land-cover change and human population trends in the greater Serengeti ecosystem from 1984-2003. Biological Conservation, 147, 255–263.

Fa, E. J., & Brown, D. (2009). Impacts of hunting on mammals in African tropical moist forests: A review and synthesis. Mammal Review, 39(4), 231-264.

94 Fa, E. J., Currie, D., & Meeuwig, J. (2003). Bushmeat and food security in the Congo Basin: Linkages between wildlife and people's future. Enviromental Conservation, 30(1), 71-78.

Fa, E. J., Olivero, J., Real, R., Farfan, M. A., Marquez, A. L., Vargas, J. M., . . . Nasi, R. (2015). Disentangling the relative effects of bushmeat availability on human nutrition in central Africa. Sci Rep, 5(8168), 1-8.

FAO. (2011). Guidelines for Measuring Household and Individual Dietary Diversity. Retrieved from Rome, Italy:

FAO. (2014). Enhancing the Contribution of Non-Wood Forest Products to Food Security in Central Africa. Retrieved from Cental Africa:

FAO. (2015). Illegal Hunting & the Bush-meat Trade in Savanna Africa: Drivers, Impacts & Solutions to Address the Problem. Retrieved from Harare, Zimbabwe:

Fischer, A., Naiman, L. C., Lowassa, A., Randall, D., & Rentsch, D. (2014). Explanatory factors for household involvement in illegal bushmeat hunting around Serengeti, Tanzania. Journal for Nature Conservation, 22(6), 491- 496.

Galvin, A. K., Polasky, S., Costello, C., & Loibook, M. (2008). Human Response to Change: Modeling Household Decision Making in Western Serengeti. In A. R. E. Sinclair, C. Packer, A. R. S. Mduma, & M. J. Fryxel (Eds.), Serengeti III: Human Impacts on Ecosystem Dynamics (pp. 325-346). Chicago and London: The University of Chicago Press.

Gavin, M. C., Solomon, J. N., & Blank, S. G. (2010). Measuring and monitoring illegal use of natural resources. Conservation Biology, 24, 89–100.

Gideon, P. K. (2014). Evaluation of wildlife hunting and species of animals marketed as bushmeat in dakka bali local government area, Taraba state, Nigeria. Journal of research in forestry, wildlife and environmental, 6(2), 66-75.

95 Golden, D. C., Bonds, H. M., Brashares, S. J., Rasolofoniaina, B. J., & Kremen, C. (2014). Economic Valuation of Subsistence Harvest of Wildlife in Madagascar. Conservation Biology, 28(1), 234-243.

Golden, D. C., Fernald, L. C., Brashares, J. S., Rasolofoniaina, B. J., & Kremen, C. (2011). Benefits of wildlife consumption to child nutrition in a biodiversity hotspot. Proc Natl Acad Sci U S A, 108(49), 19653-19656.

Golden, D. C., Rabehatonina, G. C. J., Rakotosoa, A., & Moore, M. (2014). Socio- ecological analysis of natural resource use in Betampona Strict Natural Reserve. Madagascar Conservation & Development, 9(2), 83-89.

Haule, K. S., Johnsen, F. H., & Maganga, S. (2002). Striving for sustainable wildlife management: The case of Kilombero Game Controlled Area, Tanzania. Enviromental Management, 66(1), 31-42.

Heale, R., & Twycross, A. (2015). Validity and reliability in quantitative research. Evidence-Based Nursing, 18, 66-67.

Holmern, T., Mkama, S., Muya, J., & Røskaft, E. (2006). Intraspecific prey choice of bushmeat hunters outside the Serengeti National Park, Tanzania: a preliminary analysis. African Zoology, 41(1), 81-87.

Holmern, T., Muya, J., & Røskaft, E. (2007). Local law enforcement and illegal bushmeat hunting outside the Serengeti National Park, Tanzania. Environmental Conservation, 34(1), 55-63.

Holmern, T., Nyahongo, W. J., & Røskaft, E. (2007). Livestock loss caused by predators outside the Serengeti National Park, Tanzania. Biological Conservation, 135, 518 – 526.

Holmern, T., Røskaft, E., Mbaruka, J., Mkama, S., & Muya, J. (2002). Uneconomical game cropping in a community-based conservation project outside the Serengeti National Park, Tanzania. Oryx, 36(4), 364-372.

Homans, C. G. (1958). Social Behavior as Exchange. American Journal of Sociology, 63(6), 597-606.

96 Jacob, D., Nelson, I., & Ukpong, E. (2018). Determinants of Bushmeat Traders' Income in Itu, Akwa Ibom State, Nigeria. 2, 103-116.

Kaltenborn, P. B., Nyahongo, W. J., & Kideghesho, R. J. (2011). The attitudes of tourists towards the environmental, social and managerial attributes of Serengeti National Park, Tanzania. Tropical Conservation Science, 4(2), 132- 148.

Kaltenborn, P. B., Nyahongo, W. J., & Tingstad, K. M. (2005). The nature of hunting around the Western Corridor of Serengeti National Park, Tanzania. Eur. J. Wildl. Res., 51, 213-222.

Kideghesho, J. R. (2008). Co-existence between the traditional societies and wildlife in western Serengeti, Tanzania: its relevancy in contemporary wildlife conservation efforts Biodiversity and Conservation, 17, 1861–1881.

Kideghesho, J. R. (2009). The potentials of traditional African cultural practices in mitigating overexploitation of wildlife species and habitat loss: experience of Tanzania. International Journal of Biodiversity Science & Management, 5(2), 83-94.

Kideghesho, J. R. (2010). „Serengeti shall not die‟: Transforming an Ambition into a Reality Tropical Conservation Science, 3(3), 228-248.

Kiffner, C., Peters, L., Stroming, A., & Kioko, J. (2015). Bushmeat Consumption in the Tarangire-Manyara Ecosystem, Tanzania. Tropical Conservation Science, 8(2), 318-332.

Knapp, E. J. (2007). Who Poaches? Household Economies of Illegal Hunters in Western Serengeti, Tanzania. Human Dimensions of Wildlife, 12(3), 195-196.

Knapp, E. J. (2012). Why Poaching Pays: A Summary of Risks and Benefits Illegal Hunters Face in Western Serengeti, Tanzania. Tropical Conservation Science, 5(4), 434-445.

97 Knapp, Peace, N., & Bechtel, L. (2017). Poachers and Poverty: Assessing Objective and Subjective Measures of Poverty among Illegal Hunters Outside Ruaha National Park, Tanzania. Conservation and Society, 15(1), 24-32.

Knapp, Rentsch, D., Schmitt, J., Lewis, C., & Polasky, S. (2010). A tale of three villages: Choosing an effective method for assessing poaching levels in western Serengeti, Tanzania. Oryx, 44, 178-184.

Kothari, C. R. (2004). Research Methodology: Methods and Techniques. New Delhi, India: New age international (P) Limited Publishers.

Kümpel, N. F., Milner-Gulland, E. J., Cowlishaw, G., & Rowcliffe, M. J. (2010). Incentives for Hunting: The Role of Bushmeat in the Household Economy in Rural Equatorial Guinea. Human Ecology, 38, 251 - 264.

Kyando, T. M., Nyahongo, W. J., Røskaft, E., & Nielsen, R. M. (2019). Household Reliance on Environmental Income in the Western Serengeti Ecosystem, Tanzania. Environment and Natural Resources Research, 9(1), 54-63.

Lindenberg, S. (2001). Social Rationality versus Rational Egoism. In J. Turner (Ed.), Handbook of sociological theory (pp. 635-668). New York: Kluwer Academic/Plenum.

Lindsey, J. K. (2008). Applying Generalized Linear Models. New York: Springer

Lindsey, Balme, G., Becker, M., Begg, C., Bento, C., Bocchino, C., . . . Zisadza- Gandiwa, P. (2013). The bushmeat trade in African savannas: Impacts, drivers, and possible solutions. Biodiversity and Conservation, 160, 80–96.

Loibooki, M., Hofer, H., Campbell, L. I. K., & East, L. M. (2002). Bushmeat hunting by communities adjacent to the Serengeti National Park, Tanzania: the importance of livestock ownership and alternative sources of protein and income. Environmental Conservation, 29(3), 391–398.

98 Luiselli, L., Hema, E. M., Segniagbeto, G. H., Ouattara, V., Eniang, E. A., Di Vittorio, M., . . . Fa, J. E. (2017). Understanding the influence of non-wealth factors in determining bushmeat consumption: Results from four West African countries. Acta Oecologica, 94, 47-56.

Manyama, F. F., Nielsen, R. M., Røskaft, E., & Nyahongo, W. J. (2019). The Importance of Bushmeat in Household Income as a Function of Distance from Protected Areas in the Western Serengeti Ecosystem, Tanzania. Environment and Natural Resources Research, 9(3), 49-62.

Manyama, F. F., Nyahongo, W. J., Nielsen, R. M., & Røskaft, E. (2019). Schoolchildren as informants about bushmeat consumption in Western Serengeti, Tanzania. International Journal of Biodiversity and Conservation, 11(5), 154-164.

Manyama, F.F., Nyahongo, W. J., & Røskaft, E. (2014). Factors affecting attitudes of local people toward the red-billed Quelea (Quelea quelea) in Kondoa District, Tanzania. International Journal of Biodiversity and Conservation, 6(2), 138- 147.

Mas-Colell, A., Whinston, D. M., & Green, R. J. (1995). Microeconomic Theory (Vol. 1). New York: Oxford University Press, Inc.

Masadeh, M. A. (2012). Focus Group: Reviews and Practices. International Journal of Applied Science and Technology, 2(10), 63–68.

Maseko, H., Shackleton, M. C., Nagoli, J., & Pullanikkatil, D. (2017). Children and Wild Foods in the Context of Deforestation in Rural Malawi. Human Ecology, 45, 795–807.

Masuruli, M. B. (2014). Costs and Benefits of Nature-Based Tourism to Conservation and Communities in the Serengeti Ecosystem. (PhD Thesis), University of Victoria, Canada.

99 McNamara, J., Rowcliffe, M., Cowlishaw, G., Alexander, J. S., Ntiamoa-Baidu, Y., Brenya, A., & Milner-Gulland, E. J. (2016). Characterising Wildlife Trade Market Supply-Demand Dynamics. PLoS ONE, 11(9), 1-18.

McPherson, R. S., Hoelscher, D. M., Alexander, M., Scanlon, K. S., & Serdula, M. K. (2000). Dietary Assessment Methods among School-Aged Children: Validity and Reliability. Preventive Medicine, 31(2), 11-33.

Mendelson, S., Cowlishaw, G., & Rowcliffe, J. M. (2003). Anatomy of a Bushmeat Commodity Chain in Takoradi, Ghana. Peasant Studies, 31(1), 73-100.

Mfunda, M. I., & Røskaft, E. (2010). Bushmeat hunting in Serengeti, Tanzania: An important economic activity to local people. International Journal of Biodiversity and Conservation, 2(9), 263-272.

Milner-Gulland, E. J., Bennett, E. L., & Group, A. M. W. (2003). Wild meat: the bigger picture. TRENDS in Ecology and Evolution, 18(7), 351-357.

Mmassy, E. C., Fyumagwa, R. D., Bevanger, K., & Røskaft, E. (2018). Breeding ecology of Kori Bustard Ardeotis kori strunthiunculus in the Serengeti National Park. Ostrich, 89(2), 155-162.

Morsello, C., Yagüe, B., Bestreschi, L., van Vliet, N., Adams, C., Schor, T., . . . Cruz, D. (2015). Cultural attitudes rather than economic factors are stronger predictors of bushmeat consumption and preference among urban Amazonians from Brazil and Colombia. Ecology and Society, 20(4), 21-39.

Mramba, R. P., Andreassen, H. P., & Skarpe, C. (2017). Browsing and plant traits in nutrient-rich and nutrient-poor savannas in Tanzania. Journal of Tropical Ecology, 33(5), 327-336.

Mwakatobe, A. (2013). Human-Wildlife Interaction in the Western Serengeti: Crop Raiding, Livestock Depredation and Bushmeat Utilisation. (PhD Dessertation), Norwegian University of Science and Technology, Trondheim.

100 Mwakatobe, A., Røskaft, E., & Nyahongo, W. J. (2012). Bushmeat and food security: Species preference of sundried bushmeat in communities in the Serengeti - Mara ecosystem, Tanzania. International Journal of Biodiversity and Conservation, 4(12), 548-559.

Myers, H. R., Montgomery, C. D., Vining, G. G., & Robinson, J. T. (2010). Generalized Linear Models: With Applications in Engineering and the Sciences (2 ed.). New York: John Wiley & Sons, Inc.

Nasi, R., Taber, A., & van Vliet, N. (2011). Empty forests, empty stomachs? Bushmeat and livelihoods in the Congo and Amazon Basins. International Forestry Review, 13(3), 355-368.

Ndibalema, G. V., & Songorwa, N. A. (2007). Illegal meat hunting in serengeti: dynamics in consumption and preferences. African Journal Ecology, 46, 311– 319.

Nielsen, R. M., Jacobsen, J. B., & Thorsen, J. B. (2014). Factors Determining the Choice of Hunting and Trading Bushmeat in the Kilombero Valley, Tanzania. Conservation Biology, 28(2), 382–391.

Nielsen, R. M., & Meilby, H. (2015). Hunting and trading bushmeat in the Kilombero Valley, Tanzania: motivations, cost-benefit ratios and meat prices. Environmental Conservation, 42(1), 61–72.

Nielsen, R. M., Meilby, H., & Smith-hall, C. (2014). How could the bushmeat trade in the Kilombero Valley of Tanzania be regulated? Insights from the rural value chain. Oryx, 50(1), 84–93.

Nielsen, R. M., Meilby, H., Smith-Hall, C., Pouliot, M., & Treue, T. (2018). The Importance of Wild Meat in the Global South. Ecological Economics, 146, 696-705.

101 Nielsen, R. M., Pouliot, M., & Bakkegaard, K. R. (2012). Combining income and assets measures to include the transitory nature of poverty in assessments of forest dependence: Evidence from the Democratic Republic of Congo. Ecological Economics, 78, 37-46.

Nielsen, R. M., Pouliot, M., Meilby, H., Smith-Hall, C., & Angelsen, A. (2017). Global patterns and determinants of the economic importance of bushmeat. Biological Conservation, 215, 277-287.

Nindi, S. J. (2010). Changing land-use in the fragile Lake Nyasa catchments of Tanzania: A lowland-highland nexus. Journal of Sustainable use of biological diversity in socio-ecological production landscapes, 52, 73-82.

Norton-Griffiths, M., Herlocker, D. J., & Pennycuick, L. (1975). The patterns of rainfall in the Serengeti Ecosystem, Tanzania. African Journal of Ecology, 13, 347-374.

Nuno, A., Bunnefeld, N., Naiman, L. C., & Milner-Gulland, E. J. (2013). A novel approach to assessing the prevalence and drivers of illegal bushmeat hunting in the Serengeti. Conservation Biology, 27, 1355–1365.

Nyahongo, W. J., East, M. L., Mturi, F. A., & Hofer, H. (2006). Benefits and costs of illegal grazing and hunting in the Serengeti ecosystem. Environmental Conservation, 32(4), 326-332.

Nyahongo, W. J., Holmern, T., Kaltenborn, P. B., & Røskaft, E. (2009). Spatial and temporal variation in meat and fish consumption among people in the Western Serengeti, Tanzania: the importance of migratory herbivores. Oryx, 43, 258-266.

Nyaki, A., Gray, A. S., Lepczyk, A. C., Skibins, C. J., & Dennis, R. (2013). Local- Scale Dynamics and Local Drivers of Bushmeat Trade. Conservation Biology, 00(0), 1–12.

102 Ofori, B. Y., & Attuquayefio, D. K. (2010). Hunting Intensity in the Suhuma Forest Reserve in the Sefwi Wiawso District of the Western Region of Ghana: A Threat to Biodiversity Conservation. West African Journal of Applied Ecology, 17, 135-142.

Ostrom, E. (1998). A Behavioral Approach to the Rational Choice Theory of Collective Action: Presidential Address. The American Political Science Review, 92(1), 1-22.

Primack, R. B. (2010). Essentials of Conservation Biology, Sinauer Associates Inc (5 ed.). USA.

Rentsch, D., & Damon, A. (2013). Analysis: Prices, poaching, and protein alternatives: An analysis of bushmeat consumption around Serengeti National Park, Tanzania. Ecological Economics, 91, 1-9.

Rentsch, D., Hilborn, R., Knapp, J. E., Metzger, L. K., & Loibooki, M. (2015). Bushmeat Hunting in the Serengeti Ecosystem: An Assessment of Drivers and Impact on Migratory and Nonmigratory Wildlife In R. E. A. Sinclair, L. K. Metzger, A. R. S. Mduma, & M. J. Fryxell (Eds.), Serengeti IV: Sustaining Biodiversity in a Coupled Human-Natural System (pp. 649-678). Chicago and London: University of Chicago Press.

Rentsch, D., & Packer, C. (2015). The effect of bushmeat consumption on migratory wildlife in the Serengeti ecosystem, Tanzania. Oryx, 49(2), 287–294.

Ribot, C. J. (1998). Theorizing access: Forest profit along Senegal's charcoal commodity chain. Development nd Change, 29, 308-341.

Ripple, J. W., Abernethy, K., Betts, G. M., Chapron, G., Dirzo, R., Galetti, M., . . . Young, H. (2016). Bushmeat hunting and extinction risk to the world‟s mammals. Royal Society open science, 3(160498), 1-16.

Rogan, Lindsey, A. P., & McNutt, J. W. (2015). Illegal Bushmeat Hunting in the Okavango Delta, Botswana: Drivers, Impacts and Potential Solutions Harare.

103 Rogan, Lindsey, A. P., Tambling, C. J., Golabek, K. A., Chase, M. J., Collins, K., & McNutt, J. W. (2017). Illegal bushmeat hunters compete with predators and threaten wild herbivore populations in a global tourism hotspot. Biological Conservation, 210, 233-242.

Sandelowski, M. (1995). Sample size in qualitative research. Research in Nursing & Health, 18(2), 179-183.

Schenck, M., Effa, E., Starkey, M., Wilkie, D., Abernethy, K., Telfer, P., . . . Treves, A. (2006). Why People Eat Bushmeat: Results From Two-Choice, Taste Tests in Gabon, Central Africa. Human Ecology, 34(3), 433-445.

Schulte-Herbrüggen, B., Cowlishaw, G., Homewood, K., & Rowcliffe, J. M. (2013). The Importance of Bushmeat in the livelihoods of West African cash-crop Farmers living in a faunally-depleted landscape. PLoS ONE, 8(8), 1-13.

Setsaas, T. H., Holmern, T., Mwakalebe, G., Stokke, S., & Røskaft, E. (2007). How does human exploitation affect impala populations in protected and partially protected areas? – A case study from the Serengeti Ecosystem, Tanzania. Biological Conservation, 136, 563-570.

Shively, G., Jagger, P., Sserunkuuma, D., Arinaitwe, A., & Chibwana, C. (2010). Profits and Margins along Uganda's Charcoal Value Chain. International Forestry Review, 12(3), 270-283.

Shogren, J. F., Tschirhart, J., Anderson, T., Ando, A. W., Beissinger, S. R., Brookshire, D., & Polásky, S. (1999). Why economics matters for endangered species protection. Conservation Biology, 13(6), 1257-1261.

Sinclair, A. R. E., Metzger, K. L., Mduma, S. A. R., & Fryxell, J. M. (2015). Serengeti IV: Sustaining Biodiversity in a Coupled Human-Natural System. Chicago and London: University of Chicago Press.

Sinclair, A. R. E., Packer, C., Mduma, S. A. R., & Fryxell, J. M. (2008). Serengeti III: Human Impacts on Ecosystem Dynamics. Chicago and London: University of Chicago Press.

104 Toomet, O., & Henningsen, A. (2008). Sample Selection Models in R: Package sampleSelection. Journal of Statistical Software, Articles, 27(7), 1-23.

URT. (1997a). National Environmental Policy. Dar-es Salaam, Tanzania: Vice President's Office.

URT. (1997b). National Land Policy. Dar-es Salaam, Tanzania: Ministry of Lands and Human Settlements Development.

URT. (1998). National Forest Policy. Dar-es Salaam, Tanzania: Ministry of Natural Resources and Tourism.

URT. (1999). National Tourism Policy. Dar-es Salaam, Tanzania: Ministry of Natural Resources and Tourism.

URT. (2007). The Wildlife Policy of Tanzania. Dar-es Salaam, Tanzania: Ministry of Natural Resources and Tourism.

URT. (2013). The United republic of Tanzania: 2012 Population and Housing Census Population Distribution by Administrative areas. Dar es Salaam: National Bureau of Statistics ministry of finance. van Vliet, N., & Mbazza, P. (2011). Recognizing the Multiple Reasons for Bushmeat Consumption in Urban Areas: A Necessary Step Toward the Sustainable Use of Wildlife for Food in Central Africa. Human Dimensions of Wildlife, 16(1), 45-54. doi:10.1080/10871209.2010.523924 van Vliet, N., Nabesse, C., Gambalemoke, S., Akaibe, D., & Nasi, R. (2012). The bushmeat market in Kisangani, Democratic Republic of Congo: implications for conservation and food security. Oryx, 46(2), 196–203. van Vliet, N., Nabesse, C., & Nasi, R. (2015). Bushmeat consumption among rural and urban children from Province Orientale, Democratic Republic of Congo. Oryx, 41(1), 165–174.

105 van Vliet, N., Schulte-Herbrüggen, Muhindo, J., Nebesse, C., Gambalemoke, S., & Nasi, R. (2017). Trends in bushmeat trade in a postconflict forest town: implications for food security. Ecology and Society, 22(4), 35-44.

Walelign, Nielsen, M. R., & Jacobsen, J. B. (2019). Roads and livelihood activity choices in the Greater Serengeti Ecosystem, Tanzania. PLoS ONE, 14(3), 1- 21.

Walelign, Nielsen, R. M., & Jakobsen, B. J. (2019). Price Elasticity of Bushmeat Demand in the Greater Serengeti Ecosystem: Insights for Managing the Bushmeat Trade. Frontiers in Ecology and Evolution, 7(162), 1-17.

Walsh, M. (2006). The Utilisation of Wildlife in Tanzania: Key Issues and Experiences. Retrieved from Cambridge:

Warner, L. S. (1965). Randomized Response: A Survey Technique for Eliminating Evasive Answer Bias. Journal of the American Statistical Association, 60(309), 63-69.

Wilfred, P. (2010). Towards sustainable Wildlife Management Areas in Tanzania. Tropical Conservation Science, 3(1), 103–116.

Wilfred, P. (2015). Exploring differences among illegal activities in the Ugalla Game Reserve of Western Tanzania. Tanzania Journal of Science, 41, 1-14.

Wilfred, P., Milner-Gulland, E. J., & Travers, H. (2017). Attitudes to illegal behaviour and conservation in western Tanzania. Oryx, 53(3), 513-522.

Wilkie, D., Bennett, E., Peres, C., & Cunningham, A. (2011). The empty forest revisited. Ann N Y Acad Sci, 1223(1), 120-128.

Wittek, R. (2013). Rational Choice Theory. In R. J. McGee & L. R. Warms (Eds.), Theory in Social and cultural Anthropology: An Encyclopedia (pp. 688-690): SAGE Publications.

Zar, H. J. (2010). Biostatistics Analysis (5 ed.). Upper Saddle River, New Jersey: Pearson Prentice Hall Pearson Education, Inc.

106 APPENDICES

Appendix 1: Questionnaire surveys for household meat consumption in villages adjacent the Serengeti National Park.

Abstract statement The study aims to collect information on household economic activities, availability of food resources, challenges and strategies used by local communities to overcome the challenges. Due to limited time, only a representative number of individuals aged ≥18 years will involved. The information provided will be strictly confidential and will be used solely for academic purpose.

Are you willing to participate in this survey? Yes ( ) No ( ) Questionnaire No:………… Interviewer………………………….. Date:……………Household GPS position: UTM Coordinates:………………...... Name of household head:…………………… Household rank …………………. Village name:……………………………, District:…………………...... RESPONDENT CENTERED QUESTIONS 1. Household information: (Respondent) a. Gender: Male ( ) Female ( ) b. Role in Household: HH head ( ) or other ……………………… c. Age: ………………… d. HH size ………………… e. Level of education of a respondent: No formal education ( ) Primary ( ) Secondary ( ) College ( ) University ( ) f. Occupation: Employed ( ) Unemployed ( ) Self employed ( ) ………….. g. Is any of household members employed? Yes ( ) No ( ) h. If question “g” above is Yes, please fill in this table Employment status Number of Male Number of Female Temporary Permanent Self employed Total

107

i. Religion: Christian ( ) Muslim ( ) Pagan ( ) others …………………. j. Tribe: ………………………………… k. Marital status: Married ( ) Single ( ) Divorced ( ) Window ( ) l. Name of the place of origin/birth: ………………………………… m. Number of houses………… n. Roofing material of main building; Iron sheet ( ) Thatched grass ( ) o. Wall material of building; brick and cement ( ) earth bricks ( ) earth and polls ( ) p. Does your household own the following items? Solar ( ) Radio ( ) Bicycle ( ) Motorcycle ( ) Vehicle ( ) Television ( ) Electricity ( ) others ……….. 2. Did you migrate to the current village you are residing? Yes ( ) No ( ) 3. [If Question 2 above is Yes], what was/were the reason (s) for in-migrating to this village? Marriage ( ) Employment ( ) Land for Agriculture ( ) Social services ( ) education ( ) hospital ( ) water ( ) Access to natural resources ( ) Bushmeat ( ) Forest products ( ) Tourism activities ( ) 4. If migrated to the current village, i) where did you come from? …………… ii) For how long have you been living in this village? ……………………… 5. What are the three main household economic activities (please rank them in order) i)…………………… ii)…………………… iii)…………………… 6. How much on average did you earn last month from all income sources? Tsh……. 7. What is the average monthly household income for the past 12 months? Tsh ………………………………………………………………………….. 8. Do you have any problem with wild animals? Yes ( ) No ( ) 9. If yes, what season do you face problems with wild animals? Dry season ( ) Wet season ( ) 10. What species of wild animals cause problem to your household? ……………………………………………………………………………….

108 11. Do you cultivate crops? Yes ( ) No ( ) 12. If yes, what is the size of cultivated farm? ………………………………… 13. What types of crops do you grow? Maize ( ) sorghum ( ) cassava ( ) sweat potatoes ( ) sunflower ( ) cotton ( ) beans ( ) groundnuts ( ) finger millet ( ) others ………………………………………………………………………….. 14. What stage do wild herbivores attack and destroy crops? weeding ( ) flowering ( ) harvesting ( ) all stages ( ) 15. Do you keep livestock or other domestic animals? Yes ( ) No ( ) 16. If Yes, please fill in the following table: Species Cattle Goat Sheep Poultry Donkey Pig Dog Others (specify) Number

17. How frequently do you or someone in your family participate in the following activities? Several times per week (1), several times a month (2), several times per year (3), very rarely (4), never (5) Activity Score Herding livestock Farming Fishing Hunting Tourism activities Business

109 A) Household animal meat consumption 18. What types of protein foods were eaten in your household for the past 7 days? Protein type Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Fish Sardines Chicken Sheep Pork Goat Beef Bushmeat Milk Eggs Others (specify)

19. List types of protein food which are mostly available to your household, (explain why?) Protein type Tick the Source (purchased, Reasons (low price, season, available slaughtered, others) livestock production, hunting) Fish Small fish Chicken Sheep Pork Goat Beef Bushmeat Milk Eggs Beans Peas Others

110 20. How frequently do you or someone in your family eat the following food items? Several times per week (1), several times a month (2), several times per year (3), very rarely (4), never (5). Food item Frequency Ugali Rice Makande Chicken Beef Bushmeat Fish Others (specify)

B) Bushmeat market, commodity chain and trade routes 21. Do you prefer bushmeat than other meat in your households? Yes ( ) No ( ) 22. If yes, why? Most available ( ) cheap ( ) good taste ( ) culture ( ) others ……… 23. If bushmeat are to be sold, will you be willing to buy it the same price as beef? Yes ( ) No ( ) 24. What kind of bushmeat do local people mostly prefer? Zebra ( ) Wildebeest ( ) Impala ( ) Gazelle ( ) others ………………….. 25. What kinds of animals are mostly hunted? Wildebeest ( ) Impala ( ) Gazelle ( ) Zebra ( ) Buffalo ( ) Others……….. 26. Based on your experience, who do you think participate in hunting and trading bushmeat? Young poor uneducated people ( ) Mature poor uneducated people ( ) Young poor educated people( ) Mature poor educated people ( ) Mature educated rich people ( ) Mature uneducated rich people ( ) 27. How many times does a person go for hunting within a month? ...... 28. How many animals can be hunted per one hunting trip? ……………………. 29. What is the main use of bushmeat by hunters? subsistence ( ) commercial ( ) both ( )

111 30. Currently, what is the price of bushmeat in kg? Dry meat……..… Fresh meat ……...... 31. Who sells the bushmeat in your village? Hunters ( ) Retail sellers ( ) Whole seller ( ) middle men ( ) 32. Which centers are mostly famous for bushmeat trading? ...... 33. Where does hunting take place? In open areas ( ) Protected Areas ( ) both ( ) 34. How the meat is transported from source area to other areas? By bicycle ( ) Motorcycle ( ) Car ( ) on foot ( ) by donkey ( ) 35. What are the reasons for hunting?

SN Reasons Yes No No idea If yes (explain) 1 Lack of income generating activity 2 Poverty 3 Human wildlife conflict 4 Ineffective law enforcement 5 keeping traditions 6 Cheap source of protein (food) 7 Park is close to the village 8 Quick source of cash 9 Hunting is prestigious 10 Higher prices of livestock meat 12 Others (specify)

36. How do poachers avoid being caught by park rangers?...... 37. Do you know person(s) or any member of your household who have been arrested because of illegal hunting? Yes ( ) No ( ) 38. What is the penalty when people are arrested because of illegal hunting? Fine ( ) Jailed/imprisoned ( ) others (specify) ………………………… 39. Do you know person(s) or any member of your household who currently involved in illegal hunting? Yes ( ) No ( )

112 40. What should be done in order to reduce illegal Bushmeat hunting? Increase alternative meat sources ( ) law enforcement ( ) others ……………… 41. What would make hunters and bushmeat traders shift into other livelihood strategy? Education ( ) Capital ( ) Fines/jailed ( ) others ……… 42. Do you agree that illegal hunting reduces wildlife populations and can lead to extinction? Yes ( ) No ( )

C) Seasonal influence on the supply of bushmeat 43. What kind of meat protein food is largely available all seasons? Domestic meat ( ) Bushmeat ( ) Fish ( ) poultry ( ) others…..why? ………………… 44. During this season ……... what type of meat is most available at the village? Domestic meat ( ) Bushmeat ( ) Fish ( ) poultry ( ) why? ……….. 45. Which season has the highest supply of bushmeat? wet season ( ) dry season ( ) why………………………………………….. 46. Do the prices of bushmeat change with seasons? Yes ( ) No ( ) 47. Which season has the highest price and why? Wet season ( ) Dry season ( ) Why? …………………………………………………………………….

D) Source areas of bushmeat 48. Where do you get bushmeat? From Hunters within your household ( ) purchasing ( ) 49. Who sells bushmeat to your household? Hunters ( ) retail sellers ( ) others ( ) 50. Where the bushmeat is hunted? (Explain the reason(s) for site selection) SN Reasons Yes No No idea If yes (explain why) 1 Open areas (village land) 2 Serengeti National park 3 Game reserves 4 WMAs 5 Others (specify)

Respondent status;-Hunter ( ) consumer ( ) whole seller ( ) retail seller ( ) middle men ( )

113 Appendix 2: Questionnaire surveys for hunters and bushmeat traders in villages adjacent the Serengeti National Park. Abstract statement The study aims to collect information on household economic activities, availability of food resources, challenges and strategies used by local communities to overcome the challenges. Due to limited time, only a representative number of individuals aged ≥18 years will involved. The information provided will be strictly confidential and will be used solely for academic purpose.

Are you willing to participate in this survey? Yes ( ) No ( ) Questionnaire No:………… Interviewer………………………….. Date:……………Household GPS position: UTM Coordinates:………………...... Name of household head:…………………… Household rank …………………. Village name:……………………………, District:…………………......

RESPONDENT CENTERED QUESTIONS 1. Household information: (Respondent) a. Gender: Male ( ) Female ( ) b. Role in Household: HH head ( ) or other ……………………… c. Age: ………………… d. HH size ………………… e. Level of education of a respondent: No formal education ( ) Primary ( ) Secondary ( ) College ( ) University ( ) f. Occupation: Employed ( ) Unemployed ( ) Self employed ( ) ………….. g. Is any of household members employed? Yes ( ) No ( ) h. If question “g” above is Yes, please fill in this table Employment status Number of Male Number of Female Temporary Permanent Self employed Total

114 i. Religion: Christian ( ) Muslim ( ) Pagan ( ) others …………………. j. Tribe: ………………………………… k. Marital status: Married ( ) Single ( ) Divorced ( ) Window ( ) l. Name of the place of origin/birth: ………………………………… m. Number of houses………… n. Roofing material of main building; Iron sheet ( ) Thatched grass ( ) o. Wall material of building; brick and cement ( ) earth bricks ( ) earth and polls ( ) p. Does your household own the following items? Solar ( ) Radio ( ) Bicycle ( ) Motorcycle ( ) Vehicle ( ) Television ( ) Electricity ( ) others ……….. 2. Did you migrate to the current village you are residing? Yes ( ) No ( ) 3. [If Question 2 above is Yes], what was/were the reason (s) for in-migrating to this village? Marriage ( ) Employment ( ) Land for Agriculture ( ) Social services ( ) education ( ) hospital ( ) water ( ) Access to natural resources ( ) Bushmeat ( ) Forest products ( ) Tourism activities ( ) 4. If migrated to the current village, i) where did you come from? ……………. ii) For how long have you been living in this village? ……………………… 5. What are the three main household economic activities (please rank them in order) i)…………………… ii)…………………… iii)…………………… 6. How much on average did you earn last month from all income sources? Tsh……. 7. What is the average monthly household income for the past 12 months? Tsh …….. 8. Do you have any problem with wild animals? Yes ( ) No ( ) 9. If yes, what season do you face problems with wild animals? Dry season ( ) Wet season ( ) 10. What species of wild animals cause problem to your household? ………… 11. Do you cultivate crops? Yes ( ) No ( )

115 12. If yes, what is the size of cultivated farm? ………………………………… 13. What types of crops do you grow? Maize ( ) sorghum ( ) cassava ( ) sweat potatoes ( ) sunflower ( ) cotton ( ) beans ( ) groundnuts ( ) finger millet ( ) others ……….. 14. What stage do wild herbivores attack and destroy crops? weeding ( ) flowering ( ) harvesting ( ) all stages ( ) 15. Do you keep livestock or other domestic animals? Yes ( ) No ( ) 16. If Yes, please fill in the following table:

Species Cattle Goat Sheep Poultry Donkey Pig Dog Others (specify) Number

17. How frequently do you or someone in your family participate in the following activities? Several times per week (1), several times a month (2), several times per year (3), very rarely (4), never (5) Activity Score Herding livestock Farming Fishing Hunting Tourism activities Business

116 A) Household animal meat consumption 18. What types of protein foods were eaten in your household for the past 7 days? Protein type Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Fish Sardines Chicken Sheep Pork Goat Beef Bushmeat Milk Eggs Others (specify)

19. List types of protein food which are mostly available to your household, (explain why?) Protein type Tick the Source (purchased, Reasons (low price, season, available slaughtered, others) livestock production, hunting) Fish Small fish Chicken Sheep Pork Goat Beef Bushmeat Milk Eggs Beans Peas Others

117 20. How frequently do you or someone in your family eat the following food items? Several times per week (1), several times a month (2), several times per year (3), very rarely (4), never (5). Food item Frequency Ugali Rice Makande Chicken Beef Bushmeat Fish Others (specify)

B) Hunting and trading bushmeat, commodity chain and trade routes 21. Do you prefer bushmeat than other meat in your households? Yes ( ) No ( ) 22. If yes, why? Most available ( ) cheap ( ) good taste ( ) culture ( ) others ……… 23. If bushmeat are to be sold, will you be willing to buy it the same price as beef? Yes ( ) No ( ) 24. What kind of bushmeat do local people mostly prefer? Zebra ( ) Wildebeest ( ) Impala ( ) Gazelle ( ) others ………………….. 25. What kinds of animals are mostly hunted? Wildebeest ( ) Impala ( ) Gazelle ( ) Zebra ( ) Buffalo ( ) Others……….. 26. Based on your experience, who do you think participate in hunting and trading bushmeat? Young poor uneducated people ( ) Mature poor uneducated people ( ) Young poor educated people( ) Mature poor educated people ( ) Mature educated rich people ( ) Mature uneducated rich people ( ) 27. How many times does a person go for hunting within a month? …………… 28. How many animals can be hunted per one hunting trip? ……………………. 29. How many times did you go for hunting during the past month?...... year …… 30. How much did you earn during the last hunting trip? Tsh.. from ….animals.

118 31. When was your last hunting trip? ……………… 32. How much cost did you incurred during the last hunting trip? Tsh ……other cost…… 33. What is the main use of bushmeat? subsistence ( ) commercial ( ) both ( ) 34. Currently, what is the price of bushmeat in kg? Dry meat….… Fresh meat … 35. Who sells bushmeat in your village? Hunters ( ) Retail sellers ( ) Whole seller ( ) 36. Are there middle men who connect the bushmeat trading? Yes ( ) No ( ) 37. If yes, where are they from? This village ( ) nearby villages ( ) others …… 38. Where are they found? In village centers ( ) street ( ) others (specify)……… 39. Which centers are mostly famous for bushmeat trading? ...... 40. How much income can be generated from bushmeat trading per month? … from how much Kilogram (Kg) of meat?...... 41. Who are the customers of bushmeat? People from this village ( ) people from nearby villages ( ) both ……………….. 42. If bushmeat is sold to nearby villages, mention the villages ……………………………… and route it pusses (for subsequence GPS location of the route) ……………………. 43. When compared customers from this village and nearby villages, which customers mostly buy bushmeat and why? Type of customer Yes No Reasons (Distance from PAs, livestock ownership and crop production) This village Nearby villages

44. Where does hunting take place? In open areas ( ) Protected Areas ( ) both ( ) 45. How the meat is transported from source area to other areas? By bicycle ( ) Motorcycle ( ) Car ( ) on foot ( ) by donkey ( ) 46. What are the total cost of transporting bushmeat from one area to another?..... 47. What are other costs associated with bushmeat hunting? (Opportunity cost) …………………......

119 48. What are the reasons for hunting? SN Reasons Yes No No idea If yes (explain) 1 Lack of income generating activity 2 Poverty 3 Human wildlife conflict 4 Ineffective law enforcement 5 keeping traditions 6 Cheap source of protein (food) 7 Park is close to the village 8 Quick source of cash 9 Hunting is prestigious 10 Higher prices of livestock meat 12 Others (specify)

49. How do poachers avoid being caught by park rangers?...... 50. Do you know person(s) or any member of your household who have been arrested because of illegal hunting? Yes ( ) No ( ) 51. What is the penalty when people are arrested because of illegal hunting? Fine ( ) Jailed/imprisoned ( ) others (specify) ………………………… 52. Do you know person(s) or any member of your household who currently involved in illegal hunting? Yes ( ) No ( ) 53. What should be done in order to reduce illegal Bushmeat hunting? Increase alternative meat sources ( ) law enforcement ( ) others ……………… 54. What would make people that are hunting and trading bushmeat shift into some other livelihood strategy? Education ( ) Capital ( ) Fines/jailed ( ) others ……… 55. Do you agree that illegal hunting reduces wildlife populations and can lead to extinction? Yes ( ) No ( )

120 C) Seasonal influence on the supply of bushmeat 56. What kind of meat protein food is largely available all seasons? Livestock meat ( ) Bushmeat ( ) Fish ( ) poultry ( ) others…..why? ………………… 57. During this season ……... what type of meat is most available at the village? Livestock meat ( ) Bushmeat ( ) Fish ( ) poultry ( ) why? ……….. 58. Which season has the highest supply of bushmeat? wet season ( ) dry season ( ) why………………………………………….. 59. Do the prices of bushmeat change with seasons? Yes ( ) No ( ) 60. Which season has the highest price and why? Wet season ( ) Dry season ( ) Why? ……………………………………………………………………. D) Source areas of bushmeat species 61. Where do you get bushmeat? From Hunters within your household ( ) purchasing ( ) 62. Who sells bushmeat to your household? Hunters ( ) retail sellers ( ) others ( ) 63. Where the bushmeat is hunted? (Explain the reason(s) for site selection) SN Reasons Yes No No idea If yes (explain why) 1 Open areas (village land) 2 Serengeti National park 3 Game reserves 4 WMAs 5 Others (specify)

Respondent status;-Hunter ( ) whole seller ( ) retail seller ( ) middle men ( )

121 Appendix 3: Schoolchildren Interview Guide and Data record sheet for dietary recording GPS______Village _____ District ______1. Student name ______2. Age___ 3. Sex ______4. Height (cm) ____ 5. Weight (kg)______Father‟s occupation______7. Mother‟s occupation ______8. Household size _____ 9. Physical appearance ______10. Date dewormed ______Dietary recording date started ______Date ended ______Food Monday Tuesday Wednesday Thursday Friday Saturday Sunday items B L D B L D B L D B L D B L D B L D B L D Tea Bread Cassava Sweet potatoes Fresh milk Sour milk Porridge Rice Maize mealie (Ugali) Beef Goat meat Mutton Pork Chicken Sardine Tilapia Nile perch Beans vegetable Bushmeat Makande (beans/m aize)

Key; B = Breakfast, L = lunch, D = Dinner

122 Appendix 4: Data sheet for Bushmeat Packages Data sheet No:……... Village name :…………...... Date GPS ID Weight Price Claimed Respondent Source (Kg) (Tsh) spp

123 Appendix 5: Data sheet for prices of animal meat foods Date GPS Food type Weight Price (Tsh) Remarks (Kg)

124 Appendix 6: Ethical clearance certificate for conducting research in western Serengeti

125 Appendix 7: Average market prices of different products in the study villages Product Unit Price in TZS Price in USD Bushmeat Package 4,000 1.78 Firewood Head load 3,000 1.33 Charcoal Sack (100 kg) 15,000 6.67 Thatch grass Head load 500 0.22 Poles Head load 3,000 1.33 Fodder grass Cattle/day 100 0.04 Mushroom Kg 2,000 0.89 Honey litre 8,000 3.56 Sorghum Kg 400 0.18 Finger millet Kg 900 0.40 Maize Kg 500 0.22 cattle Per individual 500,000 222.22 Sheep Per individual 60,000 26.67 Goat Per individual 50,000 22.22 Chicken Per individual 12,000 5.33 Source: Field data 2017-2018

126 Appendix 8: Publications

127