Bushmeat hunting in Serengeti, Tanzania: An important economic activity to local people

Transcription

1 International Journal of Biodiversity and Conservation Vol. 2(9), pp , September 2010 Available online ISSN X 2010 Academic Journal Full Length Research Paper Bushmeat hunting in Serengeti, Tanzania: An important economic activity to local people Iddi M. Mfunda 1 and Eivin Røskaft1 2 * 1 Department of Biology, Norwegian University of Science and Technology, NTNU), Realfagbygget, N-7491 Trondheim, Norway. 2 Centre for Advanced Study, Drammensveien, 78, 0271 Oslo, Norway. Accepted 23rd June, 2010 Unsustainable use of natural resources poses threats to conservation and livelihoods. High-levels of bushmeat hunting threaten wildlife populations and extinction of some species. This paper gives an overview of bushmeat hunting in Serengeti, Tanzania. The data on hunting was collected through interviewing 477 households in 10 villages surrounding Serengeti National Park. Our research indicates that bushmeat hunting was taking place in Serengeti and is an important economic activity in Western Serengeti. The hunting preferences differ between Western and Eastern where the latter preferred small and the former, medium-big sized wildlife. The majority depends on bushmeat as a source of protein and a few relied on it for protein and income. In Western Serengeti hunting was taking place inside the national park and game reserves and in occasions within villages. In Eastern Serengeti, hunting took place within villages. Immigration of people, ethnicity, and number of livestock were cited to influence bushmeat hunting. We suggest strengthening and widening the coverage of community based conservation outreach programs; opening doors for sustainable use; and widening the scope of benefit sharing to address household livelihoods. Strengthening law enforcement and redefining the Serengeti ecosystem are essential actions for conserving wildlife within and outside protected area networks. Key words: Bushmeat hunting, conservation, game reserves, local people, Serengeti National Park, sustainable use, Tanzania. INTRODUCTION Unsustainable use is a concern to natural resources conservation in Africa. It affects the biodiversity that provides important ecosystem services including wildlife habitats and production of food and fibre (Redmond et al., 2006). Local people appreciate the role of biodiversity in culture, diets and economy despite the difficulties to survive without much dependence on natural resources. Bushmeat is an essential source of protein to those living close to protected areas and is motivated by subsistence, commercial and cultural needs (Robinson and Bennett, 2000; Nielsen, 2006). Subsistence hunting involves a short period of hunting and the use of traditional hunting methods such as bow and arrows, snares and pitfalls (Carpaneto and Fusari, 2000; Holmern et al., 2004; *Corresponding author. roskaft@bio.ntnu.no. Tel: Holmern et al., 2006). Commercial bushmeat hunting is organized in terms of hunting gears, number of days, processing and transportation, and marketing (Leader- Williams and Milner-Gulland, 1993). The familiarity, culture and traditional inheritance increase demand for bushmeat because it provides trophies for cultural artefacts and medicinal values (Robinson and Bennett, 2000; Mockrin et al., 2005; Wilkie et al., 2005; Kideghesho, 2008). On a broader scale, bushmeat hunting is influenced by the increase in human population and poverty, weak governance, and inadequate law enforcement (Naughton-Treves, 1998; Nielsen, 2006). These factors make bushmeat hunting a conservation concern because of its adverse effects in wildlife populations and biodiversity (Hackel, 1999; Kaltenborn et al., 2005). The increase in bushmeat hunting is a cause of biodiversity loss and decline of wildlife populations in Africa (Robinson et al., 1999; Redmond et al., 2006). In

2 264 Int. J. Biodvers. Conserv. West Africa, bushmeat is used as food, commodity to trade, and play a crucial role in rituals (Wilkie and Carpenter, 1999). In Central Africa bushmeat is a source of income for local people who have limited alternative income sources. In Gabon, hunting accounts for around 15-72% of the average household income (Starkey 2004). In Eastern Africa, unsustainable bushmeat hunting is a conservation concern too (Barnett, 2000). In Kenya, bushmeat hunting occurred in 96% of the protected areas and 25% of the meat in Nairobi butcheries was bushmeat (Okello and Kiringe, 2004; Olupot et al., 2009). In Tanzania, bushmeat hunting is an important activity and threatens all categories of protected areas including the Selous Game Reserve and Serengeti ecosystem (Barrett and Arcese, 1995; Caro et al., 1998; Barnett, 2000; Holmern et al., 2002; Loibooki et al., 2002; Baldus et al., 2003). Bushmeat hunting is easily affordable for poor communities but has high impacts on wildlife populations. Unsustainable hunting has effects on ecosystem dynamics and therefore threatens the future of targeted species as well as the entire ecosystem (Apaza et al., 2002). Hunting accelerates extinctions, mostly of large mammals (Barnes, 2002). In Uganda, massive hunting reduced large mammal population by over 90% in the 1970 s (Lamprey et al., 2003). In Tanzania, illegal hunting has dramatically reduced the populations of Buffalo (Syncerus caffer) and affected populations of giraffe (Giraffe camelopardis), impala (Aepyceros melampus) and topi (Damiliscus korrigum) (Campbell and Borner, 1995; Hofer et al., 1996). In efforts to enhance conservation, most African countries adopted communitybased conservation (CBC) in 1980 s to address sustainable use, benefit sharing, and community services. The Serengeti Regional Conservation Project (SRCP) and the Community Conservation Services (CCS) of Tanzania National Parks are examples of CBC outreach programs in Tanzania and Africa. In Western Serengeti, SRCP addresses community hunting and income generating activities as incentives for collabo-\rative wildlife management. The CCS implements social oriented projects such as schools, health, and pro-vision of water to villages neighbouring Serengeti National Park. The programs had an objective to bridge the social gap between protected areas and the surrounding local people by improving relationship and facilitating the sharing of benefits from conservation (Mfunda et al., Norwegian University of Science and Technology, Trondheim, personal communication). The aim of this paper was to review the level of illegal bushmeat hunting in Western and Eastern Serengeti of northern Tanzania. The specific objectives of the study was to: (i) assess the causes of bushmeat hunting, (ii) evaluate the effects of CBC on bushmeat hunting, (iii) identify the link between crop loss and livestock depredation with bushmeat hunting, and (iv) develop recommendations on how to control unsustainable bushmeat hunting. We compared different ethnic tribes in the three districts (Bunda and Serengeti in Western Serengeti and Ngorongoro in Eastern Serengeti); assessed the magnitude of bushmeat hunting, and hypothesized that tribes that are traditional hunters are more involved in bushmeat hunting. Further we hypothesized that the agro-pastoralists and pastoralists would risk grazing live-stock inside national park and game reserves to facilitate bushmeat hunting. This paper will provide knowledge on the importance of bushmeat hunting to local people, the understanding of the dynamics of bush-meat hunting to protected area managers, and inputs to enhance con-servation and livelihoods. METHODS Study area The study was conducted in ten villages in Bunda, Serengeti, and Ngorongoro districts in Western and Eastern Serengeti (Figure 1). The study villages borders and serves as migratory routes of wildlife to and from: Serengeti National Park (SNP) ( km 2 ) and Ngorongoro Conservation Area (NCA) (8 288 km 2 ) which are World Heritage Sites and Biosphere Reserves; Ikorongo-Grumeti Game Reserves (GR) (563 and 416 km 2, respectively); Loliondo Game Controlled Area (GCA) (4,000 km 2 ); and Ikona wildlife management area (WMA) (242 km 2 ) (Table 1). The study area contains a very high diversity and concentrations of ungulates, large carnivores, and birds (Sinclair and Arcese, 1995). The wildlife of Serengeti are migratory in nature and dominated by wildebeest (Connochaetes taurinus), zebra (Equus burchelli), thomson gazelle (Gazella thomsoni) and other threatened or endangered species like African elephant (Loxodanta Africana) (Sinclair and Arcese, 1995; Thirgood et al., 2004). The SNP is managed mainly for ecosystem protection and recreation; the Ngorongoro Conservation Area allows tourism, settlement and livestock keeping by Maasai pastoralists. The game reserves restrict any form of human activities including hunting and grazing of livestock. In the GCAs only wildlife consumption is regulated but other forms of land use including grazing of livestock, cultivation or human settlements are not restricted (URT, 1974). In the GRs and GCAs the resident hunting is allowed under the written permission of the director of wildlife (URT, 1974; Baldus and Cauldwell, 2004). In contrast, the tourist hunting regulation restricts other forms of hunting in areas designated as tourist hunting blocks (URT, 2002). Thus the Western Serengeti had no provision for residence and community hunting because Ikorongo-Grumeti GRs and Ikona WMA (the former Ikoma Open Area) are legal hunting concessions of a private company (Mfunda and Røskaft, Norwegian University of Science and Technology, Trondheim, personal communication). The human population of the three districts is estimated to be in 2002 (URT, 2002). The population densities varied among districts where Bunda had the highest (91.3 persons per km 2 ) followed by Serengeti (16.1 persons per km 2 ) and Ngorongoro (9.6 persons per km 2 ) (URT, 2002). The study villages comprise 13 ethnic tribes including Maasai, Ikoma, Sukuma, Ikizu, Natta, and Taturu. People in the study villages cultivate crops such as sorghum, finger millet, maize, cassava, sweet potatoes and beans. The livestock keeping is an important activity in Eastern Serengeti by the Maasai who are pastoralists, and in Western Serengeti by other ethnic tribes who are agro-pastoralists and they keep cattle, goats, sheep, donkeys, dogs, and poultry. The ethnic tribes were grouped into: (i) Ikoma and Natta (hunter-gatherers), (ii) Sukuma, Taturu and Ikizu (agro-pastoralists), and (iii) the Maasai (pastoralists) based on the ethnic composition, economic pontentialities, and cultural dominance. As a result the social structure

3 Mfunda and Røskaft 265 Figure 1. Map of the study area showing: Serengeti National Park, Ngorongoro Conservation Area, Ikorongo and Grumeti Game Reserves, and Loliondo Game Controlled Area; Bunda, Serengeti and Ngorongoro Districts in the West and Eastern of Serengeti National Park (); the study villages earmarked by a symbol () and the households surveyed denoted by a symbol ( ). The thick lines illustrate protected areas boundaries. Table 1. Results of bivariate analyses of households involvement in hunting ( yes or no) between ethnic groups, number of livestock owned of immigration status of people) in Western and Eastern Serengeti (N = sample size, percentage, differences tested with 2 tests, NS, not significant). Variables (N = 477) Ethnic groups (%) (%) Ikoma and Natta tribes 10.4 (18) 89.6 (155) Ikizu, Sukuma and Taturu tribes 4.9 (7) 95.1 (136) Maasai 13.7 (22) 86.3 (139) Involvement in hunting Final logistic model Yes No 2 df P = Number of livestock owned Poor ( 50) 7.4 (25) 92.6 (311) Rich ( 50) 15.6 (22) 84.4 (119) Immigration status of people Yes 5.1 (9) 94.9 (169) No 12.7 (38) 87.3 (261) = < 0.007

4 266 Int. J. Biodvers. Conserv. reflected a traditional pattern, for instance, the hunter-gatherers and agro-pastoralists lived in Western Serengeti, and the Eastern Serengeti was largely inhabited by the pastoralists. Data collection The data on bushmeat hunting was collected through questionnaire surveys in ten villages from Bunda, Serengeti and Ngorongoro districts. The interviews were conducted from January to December, 2008 and covered 477 households who were randomly selected for interview. We interviewed household heads or their wives or resident adults ( 18 years). The villages and sub-villages were picked based on a random-systematic selection. The households to be interviewed were randomly selected from the village and sub-village registers. In terms of gender 44.7% of the interviewed respondents were females and 55.3% were males, reflecting a gender consideration. The data were collected by the main researcher, a research assistant, and field assistants conversant with the village and households, languages, and culture. The questions were close-ended and open-ended questions aimed at extracting respondent opinion in an open minded atmosphere. The questionnaire addressed socio-demographic variables, crop loss from wildlife and livestock depredation, participation in wildlife management, and bushmeat hunting. The questions specific to this study addressed: (i) households involvement in hunting (yes, no), (ii) hunting preferences (small sized wildlife, medium-big sized wildlife), (iii) reasons for hunting (protein, protein and income), and (iv) hunting places (national park, game reserves, village areas,). Illegal bushmeat hunting is a sensitive issue to households. Therefore participant observation and information from secondary sources were applied to complement the information we received from the respondents. The Western Serengeti villages were involved in the CBC outreach scheme under the Serengeti Regional Conservation Project (SRCP). The community hunting was introduced in 1993 to 14 villages in Bunda and Serengeti project villages (Holmern et al., 2002). A quota for community hunting was set and endorsed by the Wildlife Division at the request of SRCP on behalf of local people. The selling of meat was confined to the project villages at a price set by villagers themselves. The CCS of Tanzania National Park covered all villages surrounding the Serengeti National Park. The presence of SRCP and SNP set precedence for comparisons. Statistical analyses Statistical tests were conducted using Statistical Package for Social Sciences (SPSS, 16). Chi-square tests were applied to tests for the differences in the dependent variables: involvement in hunting, hunting preferences, reasons for hunting, and hunting places. A binary logistic regression was used to identify the effects of sociodemographic factors independently and the cumulative effects of all variables on the independent variables. The independent variables included the ethnic background, age of respondents, gender (females and males), household size, education (educated and uneducated), number of livestock (few and many), crop loss from wildlife, livestock depredation, and immigration status of people. The exercise facilitated a chi-square test to determine if the ethnic background could influence the dependent variables. We further categorized livestock holdings into poor ( 50) and rich ( 50) and a cow was equivalent to four goats or sheep. In analyses we pooled the bushmeat hunting places into SNP and GRs because of low sample size and the similarities between the national park and game reserves in terms of law enforcement. The dependent variables inter-correlated between 0.031< rho < indicating a weak and modest correlation. We defined a household as a group of persons who live together and share expenses. We further defined the complex Maasai households as a group of people sharing dwelling houses or encampment, claim a kinship relationship, sharing responsibilities for managing a communal herd and may or may not eat from a common pot, but they are under the authority of one person who is the head of household (Kisoza, 2007). Only statistically significant results are presented and discussed. RESULTS Involvement in hunting The difference in frequencies of households claimed to be involved in bushmeat hunting between the three ethnic groups (see study area) was statistically significant (Table 1). Overall, the majority of respondents (90.1%) claimed not to be engaged in bushmeat hunting while few households admitted to conduct hunting (9.9%). The immigration status of people (immigrated people more frequent involved), and the number of livestock owned (people with few livestock less frequently involved) significantly influenced a household s involvement in hunting (Table 1). The result from a logistic regression analysis indicated that the number of livestock owned and immigration status of people was the variables having a significant effect and explained 4.8% of the variation in household involvement in bushmeat hunting (Table 5). Hunting preferences There was a significant difference between the three ethnic groups on bushmeat hunting preferences (Table 2). The ethnic groups in Western Serengeti preferred medium-big sized wildlife such as wildebeest, zebra and buffalo (75.8 and 55.8%, respectively = 66.7%), while the Maasai of Eastern Serengeti (100%) preferred small sized wildlife like impala (Aepyceros melampus), Thomson gazelle and Grants gazelle (Gazella granti). Education status and number of livestock owned significantly influenced hunting preferences (Table 2). The educated and those who had few livestock preferred medium-big sized wildlife. In contrast, people who had no education and households that had many livestock preferred small sized wildlife (Table 2). The result from the logistic regression analysis indicated that ethnic background was the only important variable in explaining bushmeat hunting preferences and explained 21.1% of the variation (Table 5). Reasons for hunting The reasons for being engaged in bushmeat hunting significantly varied among the ethnic groups (Table 3). The majority of respondents perceived to hunt in order to fulfil their protein demands (75.2%) compared to a few

5 Mfunda and Røskaft 267 Table 2. Results of bivariate analyses of hunting preferences (small sized or medium-big sized wildlife) between different ethnic groups, education status and number of livestock owned in Western and Eastern Serengeti (N = sample size, percentage, differences tested with 2 tests, NS, not significant). Hunting preferences Final logistic model Variables (N = 150) Small sized wildlife Medium-big sized wildlife 2 df P Ethnic groups (%) (%) Ikoma and Natta tribes 34.8 (23) 65.2 (43) Ikizu, Sukuma and Taturu tribes 31.7 (19) 68.3 (41) Maasai 100 (24) < Education status Uneducated 64.3 (18) 35.7 (10) Educated 39.3 (48) 60.7 (74) Number of livestock owned Poor ( 50) 37.8 (45) 62.2 (74) Rich ( 50) 67.7 (21) 32.3 (10) = = Table 3. Results of bivariate analyses of the reasons for hunting (proteins or proteins and income) between different ethnic groups, gender, education level, number of livestock owned or crop loss in Western and Eastern Serengeti (N = sample size, percentage, differences tested with 2 tests, NS, not significant). Reasons for hunting Final logistic model Variables (N = 153) Protein Protein and income 2 df P Ethnic groups (%) (%) Ikoma and Natta tribes 65.2 (45) 34.8 (24) Ikizu, Sukuma and Taturu tribes 76.7 (46) 23.3 (14) Maasai tribe 24 (100) < Gender Females 85.5 (53) 14.5 (9) Males 68.1 (62) 31.9 (29) Education status uneducated 92.9 (26) 7.1 (2) Educated 71.2 (89) 28.8 (36) = = who were motivated by protein and income (24.8%). Gender (females more motivated in protein), education level (uneducated being more concerned with protein) had a significant effect on household motivations to hunt (Table 3). The results from the logistic regression indicated that gender and ethnic background explained 18.8% of the variation in the household motivations to be engaged in hunting (Table 5). Hunting places The bushmeat hunting places differed significantly between the three ethnic groups (Table 4). The Ikoma and Natta, and Sukuma and others hunted inside the national park and game reserves (57 and 83%, respectively) while the Maasai households reported to hunt within their village areas (78%). Education level, and number of livestock owned had a significant influence on bushmeat hunting places. The uneducated and households with many livestock claimed hunting within village areas (Table 4). The results from the logistic regression analysis indicated that the number of livestock owned was the only significant variable explaining the variation in the selection of hunting places, and explained 7.3% of the variation (Table 5). DISCUSSION Involvement in hunting Generally few households are portrayed to be involved in bushmeat hunting. When asked if any member of the

6 268 Int. J. Biodvers. Conserv. Table 4. Results of bivariate analyses of hunting places between different ethnic groups, education levels and number of livestock owned in Serengeti hunting places (national parks, or game reserves; N = sample size, percentage, differences tested with 2 tests, NS, not significant). Hunting places Final logistic model Variables (N = 136) National park and game reserves Within villages 2 df P Ethnic groups (%) (%) Ikoma and Natta tribes 56.9 (37) 43.1 Ikizu, Sukuma and Taturu tribes 83.0 (44) 17.0 (9) Maasai tribe 22.2 (4) 77.8 (14) < Education Uneducated 40.9 (9) 59.1 (13) Educated 66.7 (76) 33.3 (38) Number of livestock owned Poor ( 50) 67.9 (76) 32.1 (36) Rich ( 50) 37.5 (9) 62.5 (15) = < Table 5. The effects of age, gender, household size, education, number of livestock, and ethnic background on hunting, hunting preferences and places, reasons for hunting and sources of bushmeat to households (N = sample size, *) P < 0.05, **) P < 0.01, ***) P < 0.001). Independent variables Involvement in hunting (N = 477) Hunting preferences (N = 150 ) Reasons for hunting (N = 153) Hunting places (N = 136) t-value t-value t-value t-value Ethnic background *** 11.5** 2.4 Age (in years) Gender ** 0.4 Household size Education status Number of livestock 7.5* ** Crop loss Livestock depredation Immigration status 7.4** r *** 0.288*** 0.255*** 0.125*** household was involved in bushmeat hunting, only 10% confessed. When asked about the sources and the bushmeat hunting preferences, 31% answered positively. Therefore we concluded that bushmeat hunting was taking place in Western and Eastern Serengeti although at different scales. This corroborates with other research, which establish that the numbers of meat meals are higher in the villages close to SNP boundaries (Nyahongo et al., 2009). Also a survey conducted in Bunda and Serengeti districts in 2001 found that nearly 30% of households were being involved in bushmeat hunting (Holmern et al., 2004; Johannesen, 2006). Most of those who denied eating bushmeat were considered to fear legal interventions. Our results disclosed that local people do not share the authorities definition of an illegal bushmeat hunter or a poacher. People identified as poachers are perceived as hunters. The term poacher represents the criminalization of traditional bushmeat hunters through legal construction (URT, 1974; Honey, 2008). For this reason, the collected data needs to be treated cautiously, because we may have been lacking important information due to fear from respondents. The application of questionnaire survey on bushmeat hunting could read to biases in the data analyses. The degree of explanation by the independent variables was low but showed significant results. Hence, suggesting that the low variation to variables may be due to fact that most respondents denied being involved in bushmeat hunting, and likewise their responses to hunting preferences, reasons for hunting and the hunting places. The results further suggest that wildlife motivated immigration of people into villages closer to the national park and game reserves. The movement of people were

7 Mfunda and Røskaft 269 more eminent in the Western Serengeti. As a result, the Western Serengeti is heavily populated and is becoming more multi-ethnic and multi-cultural. The high human population density is reported to be accelerated by the natural growth rate (2.9%), agriculture, livestock grazing, and hunting (Kauzeni and Kiwasila, 1994; URT, 2002; Thirgood et al., 2004). Hunting preferences The study indicates that the ethnic groups in Western Serengeti specialized in medium-big sized wildlife and the Maasai liked the small sized wildlife. This study associated the hunting preferences with the migration pattern of wildlife. When migrating to and from Serengeti ecosystem, the wildlife pass through villages before entering SNP and Ikorongo-Grumeti GR. This situation makes the choice easy because wildlife are plenty and simple to hunt in the absence of effective law enforcement in villages and to the migratory routes. A similar pattern is reported by other researches (Arcese et al., 1995). Further, other researches identified common species such as buffalo and wildebeest as most preferred species by bushmeat hunters in Western Serengeti (Holmern et al., 2006; Ndibalema and Songorwa, 2007). Some preferences are considered by this study to be associated with trading of bushmeat, because the main reasons given for hunting were suggested to be protein and income (24.8%). The Eastern Serengeti households frequently hunted small sized wildlife because of being more available inside Loliondo GCA as resident species. In Eastern Serengeti, hunting was considered to be more of a hobby than a food or income generating activity. The strong influence of the ethnic background suggested an existence of cultural attachment by different ethnic tribes to wildlife. A similar research in Western Serengeti documented differences in consumption and bushmeat preferences among ethnic tribes (Ndibalema and Songorwa, 2007). Most households in Western Serengeti do not eat bushbuck (Tragelaphus scriptus) because it is believed to cause leprosy, and is therefore culturally forbidden (Mfunda, 2001; Kideghesho, 2008). Likewise the consumption of roan antelope (Hippotragus equinus) is forbidden because killing it would bring bad luck and possibly death to the family (Mfunda, 2001). This study perceived these cultural attachments as opportunities that exist to foster sustainable use through the application of regulatory instruments to traditional resource management systems. It is important however to note that cultural attachment is not sufficient to conserve wildlife because of its limitations. Gender, education status, and number of livestock owned had an insignificant effect on hunting preferences. In this study we considered the hunting preferences to be as well associated with the taste, availability of residence and migratory species, and the objectives of hunting. Our explanations are largely supported by other studies in Serengeti and Africa (Kaltenborn et al., 2006; Hoffman, 2008). Reasons for hunting The ethnic groups in Western and Eastern Serengeti identified the need for protein as the main reason for bushmeat hunting. According to Holmern et al. (2002) the bushmeat hunters in Western Serengeti were hunting for their own consumption. Nevertheless, 30% of the respondents in Western Serengeti identified both protein and income as the main reasons for hunting. When comparing the three ethnic groups, more tribes in Western Serengeti claimed hunting to obtain protein (Tables 4), implying higher bushmeat hunting in Serengeti and Bunda, and consequently in Western Serengeti (Kideghesho, 2008). The results support our hypothesis that traditional hunter-gatherers are more involved in hunting. The findings from this study suggest the low productivity in pastoral and agro-pastoral production systems and high poverty in the study area. As pointed out in the results, females were more concerned with protein while males, as heads of households, were focusing on protein and income (Tables 3 and 5). Our findings are supported by other studies in Serengeti, which indicated that most local people consider bushmeat to be a source of protein and means of generating income (Barnett, 2000; Campbell et al., 2001; Holmern et al., 2002). The study appreciated the fact that in the Eastern Serengeti there are obvious changes in the diet of the Maasai towards non-livestock food and animal products. The history of people and conservation in Serengeti explained the cultural attachment to bushmeat hunting by different ethnic groups. The SNP and Ikorongo-Grumeti GRs occupy lands that were once occupied by people, livestock and cultivation (Neumann, 1998; Igoe, 2004; Kideghesho et al., 2007). Most tribes viewed the protected areas as their original lands and they assumed to have the right to use them. From this historical perspective, the local people do not believe in the transformations and relocations and instead they continued carrying out activities such as hunting, collection of medicinal plants, and grazing inside the areas (Mfunda, 2001). In this study we neither found any relation between crops and livestock losses from wildlife, nor any relation between the number of livestock owned and bushmeat hunting. Our understanding from this precedent is that to some villages such as Robanda, Mugeta and villages in Eastern Serengeti, crop production was considered as less important activity. This meant that crop loss from wildlife and livestock depredation derived hunting is at a lower level than the need for bushmeat itself. The households that did not report crop loss from wildlife

8 270 Int. J. Biodvers. Conserv. (30%) were much more dependent on bushmeat for food and income. In agro-pastoralists and pastoralist societies, livestock represent a bank stock to be purchased and sold when the need arises. In this regard, livestock can not readily be consumed in areas like Serengeti where other sources of proteins were plentiful. Our results indicated that people with many livestock (87.1%) relied more on bushmeat than those with few livestock (72.1%). Bushmeat was easily available and cheap compared to beef and fish. In Mugeta and Robanda, the price of one portion (estimated to one kilogramme) of bushmeat was 50-64% cheaper than the price of one kilogramme of beef. Our observation is similar to the findings of Holmern et al. (2002), that fresh bushmeat from SRCP s community hunting was cheaper ($ per kg) than beef (US $ 1.1 per kg). Thus this study suggests that the causes for hunting should not be limited to poverty or cultural attachment, but rather to other factors such as number of livestock, taste and affordability because of their chances to influence hunting in various smart ways. As reported in other research findings, in pastoral and agro-pastoral societies, wealth is traditionally reflected by the herd size and family rather than in material possessions (Hill, 2004; Sachedina, 2008). Hunting places In Western Serengeti, bushmeat hunting was taking place inside the national park and game reserves, and on occasions, such as during migration periods, it occurred within village areas. Villages like Robanda, Nyichoka, and Nyakitono/Makundusi are situated within wildebeest migratory routes. The large herds of wildebeests use non-protected areas while migrating to northern Serengeti (Rusch et al., 2005). As a result, the hunters would be using the villages and the buffer zones as their hunting destinations. The study on human-wildlife conflicts and hunting disclosed that in Western Serengeti some hunters go into protected areas to hunt and others hunt within or close to village areas (Holmern et al., 2004). In Eastern Serengeti, hunting occurred within village areas because 49% of Ngorongoro district falls within Loliondo GCA where human activities and wildlife co-exists (Mfunda and Røskaft, Norwegian University of Science and Technology, Trondheim, personal communication). This finding confirms other reports in Serengeti that bushmeat hunting is largely dependent on migratory herds (Holmern et al., 2002). This being the case, the seasonality may as well influence the choice of hunting places. Number of livestock owned had an influence on the choice of hunting places. People with few livestock were more frequently taking the risk of hunting inside the national park and game reserves. This do not support the hypothesis that people with more livestock would be more likely take a chance to hunt inside the national park and game reserves. Our results contradicts other findings from the same area that involvement in bushmeat hunting decreases with increase in the number of livestock owned (Loibooki et al., 2002). Both households with many and few number of livestock were involved in bushmeat hunting at different scales and locations. CBC and bushmeat hunting The presence of bushmeat hunting in the Western Serengeti challenges the successes of SRCP and CCS in achieving conservation and human development aspects through participatory management. The community wildlife hunting scheme by SRCP was terminated and instead the hunting was being sold to private investor (Mfunda et al., Norwegian University of Science and Technology, Trondheim, personal communication). Other findings reveal that the hunting quotas were not fully utilized, which denied local people of both bushmeat and income (Holmern et al., 2002; Bryceson et al., 2005). Weather conditions, migratory behaviour of hunted species, and inadequate capacity were among the reasons for unsatisfactory performance of the project (Mfunda 2001; Holmern et al., 2002). It is further reported by other studies that CBC outreach activities have limited capacity to reduce illegal bushmeat hunting (Nielsen, 2006; Kaltenborn et al., 2008). The risks associated with heavy dependency on development partners and conflict of interests between stakeholders may have contributed to the termination of the community hunting scheme. Nonetheless, sustainable use remains a key component of community based conservation. CONCLUSIONS AND RECOMMENDATIONS This study has shown that bushmeat hunting is an important economic activity in the Western Serengeti and is slowly picking up in the Eastern Serengeti. It is further concluded that bushmeat hunting influenced immigration of people to villages close to the national park and game reserves, particularly in the Western Serengeti. Bushmeat hunting was largely driven by ethnic background of tribes and the number of livestock owned. The low productivity in pastoral and agro-pastoral production systems increased the dependence on wildlife for food and income and translates the high poverty in the study area. The hunting places are determined by seasons, socio-economic activities (e.g. crop production and livestock grazing) and the bushmeat availability. This study further concludes that bushmeat hunting preferences differed between Western and Eastern Serengeti. The former preferred medium-big sized wildlife and the latter favoured small sized wildlife. The hunting preferences are mostly influenced by the cultural attachment embedded to wildlife, and the migration pattern of

9 Mfunda and Røskaft 271 wildlife particularly in Western Serengeti. From the findings, it is concluded that hunting is mostly done to fulfil protein needs and demands for income to meet household s basic needs. These factors however, do not apply in Eastern Serengeti where hunting is more of a hobby than a livelihood activity. The bushmeat hunting is taking place inside the national park, game reserves, and within villages. Therefore it suffices to conclude that, bushmeat hunting remains an important activity in Western Serengeti and involves a wide scale of hunting within and beyond village boundaries. From this viewpoint, the study recommends the strengthening and widening the coverage of communitybased conservation programs; opening doors for sustainable use to address poverty and cultural aspects in a holistic manner; and widening the scope of benefits from conservation by reflecting resources directly linked to peoples livelihoods. The investment in community services would reduce unsustainable use and improve crop production and livestock keeping. We also recommend the strengthening of law enforcement and improve protection of wildlife migratory routes and the buffer zones. Finally, we recommend the redefining of the Serengeti-Mara migratory ecosystem to encompass the areas outside the core protected area (e.g. national parks and game reserves) to play a crucial role in conservation (e.g. wildlife corridors and dispersal areas). ACKNOWLEDGEMENTS We appreciate the financial support from the Ministry of Natural Resources and Tourism of Tanzania and the Royal Norwegian Embassy (Dar Es Salaam, Tanzania). We are grateful to Tanzania Wildlife Research Institute, Tanzania National Parks, and the respective District Councils) for their supports. We highly recognize the valuable comments of the two anonymous reviewers that significantly improved the quality of the paper. REFERENCES Apaza L, D Wilkie E, Byron T, Huanca W, Leonard E, Pérez, V, Reyes- García V, Vadez RG (2002). Meat prices influence the consumption of wildlife by the Tsimane'Amerindians of Bolivia, Oryx. 36: Arcese P, HandoJ, Campbell K (1995). Historical and present-day antipoaching efforts in Serengeti. Serengeti II. Dynamics, managements, and conservation of an ecosystem. ARE Sinclair and P Arcese. Chicago, Ill, USA, University of Chicago Press: pp Baldus R, Kibonde B, Siege L (2003). Seeking conservation partnerships in the Selous Game Reserve, Tanzania, Parks 13: Baldus RD, Cauldwell AE (2004). Tourist hunting and its role in development of wildlife management areas in Tanzania. Sixth International Game Ranching Symposium. Paris, France. Barne RFW (2002). The bushmeat boom and bust in West and Central Africa., Oryx. 36: Barnett R (2000). Food for thought: the utilization of wild meat in Eastern and Southern Africa. Nairobi, Kenya, TRAFFIC/WWF/IUCN. Barrett CB, Arcese P (1995). Are intergrated conservation-development projects (ICDPs) sustainable, On the conservation of large mammals in Sub-Saharan Africa." World Dev. 23: 12. Bryceson I, Havnevik K, Isinika A, Jørgensen I, Melamari L Sonvinsen S (2005). Management of natural resources programme; Midterm Review of TAN 092 Phase III ( ). Final Report. Dar es Salaam, Tanzania, Ministry of Natural Resources and Tourism and The Royal Norwegian Embassy. Campbell K, Borner M (1995). Population trends and distribution of Serengeti herbivores: implications and management. Serengeti II: Dynamics, management and conservation of an ecosystem. ARE Sinclair and P Arcese. Chicago, Ill, University of Chicago Press pp Campbell KLI, Nelson V, Loibooki B (2001). Sustainable use of wildland resources: ecological, economic and social interractions: an analysis of illegal hunting of wildlife in Serengeti National Park, Tanzania. Natural Resources Institute, Final Technical Report. Chatham, Kent, UK, Department for International Development (DFID) Animal Health Programme and Livestock Production Programmes p. 56. Caro TM, Pelkey N, Borner M, Campbell KLI, Woodworth BL, Farm BP, Kuwai JO, Huish SA, Severre ELM (1998). Consequences of different forms of conservation for large mammals in Tanzania: preliminary analyses, Afr. J. Ecol. 36: Carpaneto GM, Fusari A (2000). Subsitence hunting and bushmeat exploitation in Central-western Tanzania, Biodiv. Conserv. 9: Hackel JD (1999). Community conservation and the future of Africa's wildlife." Conserv. Biol. 13: Hill CM (2004). Farmers' perspectives of conflict at the wildlifeagriculture boundary: some lessons learned from African subsistence farmers, Hum. Dimen. Wildl. 9: Hofer H, Campbell KLI, East ML, Huish SA (1996). The impact of game meat hunting on target and non-target species in the Serengeti. The exploitation of mammal populations. J Taylor and N Dunstone. London, UK, Chapman and Hall pp Hoffman LC (2008). The yield and nutritional value of meat from African ungulates, camelidae, rodents, ratites and reptiles, Meat. Sci. 80: Holmern T, Johannesen AB, Mbaruka J, Mkama SY, Muya J, Røskaft E (2004). Human-wildlife conflicts and hunting in the western Serengeti, Tanzania. NINA Project Report Trondheim, Norway, Norwegian Institute of Nature Reseach 26: 26. Holmern T, Mkama SY, Muya J, Røskaft E (2006). Intraspecific prey choice of bushmeat hunters outside the Serengeti National Park, Tanzania: A preliminary analysis. Afr. Zool. 41: Holmern T, Røskaft E, Mbaruka J, Mkama SY, Muya DJ (2002).Uneconomical game cropping in a community-based conservation project outside the Serengeti National Park, Tanzania. Oryx. 36: Honey M (2008). Ecotourism and sustainable development: who owns paradise? Washington DC, USA, Island Press. Igoe J (2004). Conservation and globalization: A study of national parks and indigenous communities from East Africa to South Dakota. Denver, USA, University of Colorado. Johannesen AB (2006). Designing intergrated conservation and development projects. Environ. Dev. Econ. 11: 20. Kaltenborn BP, Bjerke T, Nyahongo JW, Williams DR (2006). Animal preferences and acceptability of wildlife management actions around Serengeti National Park, Tanzania, Biodiv. Conserv. 15: Kaltenborn BP, Nyahongo JW, Kideghesho JR Haaland H (2008.) Serengeti National Park and its neighbours - do they Interact. J. Nat. Conserv. 16: Kaltenborn BP, Nyahongo JW, Tingstad KM (2005). The nature of hunting around the western corridor of Serengeti National Park, Tanzania. Eur. J. Wildl. Res. 51: Kauzeni AS, Kiwasila HL (1994). Serengeti regional conservation strategy: a socio economic study. Dar es Salaam, Tanzania, Institute of Resource Assessment, University of Dar Es Salaam. Kideghesho J (2008). Co-existence between the traditional societies and wildlife in western Serengeti, Tanzania: its relevancy in contemporary wildlife conservation efforts. Biodiv. Conserv. 17: Kideghesho JR, Røskaft E, Kaltenborn BP (2007). Factors influencing conservation attitudes of local people in Western Serengeti,

10 272 Int. J. Biodvers. Conserv. Tanzania. Biodiv. Conserv. 16: Kisoza LJA (2007). The role of local institutions in the management of agro-pastoral systems: A case study of Mkata plains, Kilosa district and Ngorongoro Conservation Areas, Ngorongoro district, Tanzania.. Forestry Department. Morogoro, Tanzania, Sokoine University of Agriculture. PhD. Lamprey R, Buhanga E, Omoding J (2003). A study of wildlife distributions, wildlife management systems, and options for wildlifebased livelihoods in Uganda. Kampala, Uganda, IFPRI/USAID. A report for IFPRI (International Food Policy Research Institute)/USAID. Leader-Williams N, Milner-Gulland EJ (1993). Policies for the enforcement of wildlife laws: the balance between detection and penalties in Luangwa Valley, Zambia." Conserv. Biol. 7: Loibooki M, Hofer H, Campbell KLI, East ML (2002). Bushmeat hunting by communities adjacent to the Serengeti National Park, Tanzania: the importance of livestock ownership and alternative sources of protein and income. Environ. Cons. 29: Mfunda, IM (2001). Socio-economic incentives for community-based wildlife management in the Serengeti ecosystem, Tanzania Forestry Department. Morogoro, Tanzania, Sokoine University of Agriculture. MSc (Forestry). Mockrin MH, Bennett EL, La Bruna DT (2005). Wildlife farming: A viable alternative to hunting in tropical forests? New York, USA, Wildlife Conservation Society. WCS Working Paper p. 23. Naughton-Treves L (1998). Predicting patterns of crop damage by wildlife around Kibale National Park, Uganda.Conserv. Biol. 12: Ndibalema VG, Songorwa AN (2007). Illegal meat hunting in Serengeti: dynamics in consumption and preferences, Afr. J. Ecol. 46: Neumann RP (1998). Imposing wilderness: struggle over livelihood and nature preservation in Africa. Berkely, NJ, USA, University of California Press. Nielsen MR (2006). Importance, cause and effect of bushmeat hunting in the Udzungwa mountains, Tanzania: implications for communitybased wildlife management. Biol. Conserv. 128: Nyahongo JW, Holmern T, Kaltenborn BP, 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: Okello MM, Kiringe JW (2004). Threats to biodiversity and their implications in protected and adjacent dispersal areas of Kenya. J. Sust. Tour. 12: Olupot W, Alastair JM, Andrew JP (2009). An analysis of socioeconomics of bushmeat hunting at major hunting sites in Uganda WCS Working Paper No. 38. New York, USA, Wildlife Conservation Society. Redmond I, Aldred, T, Jedamzik K, Westwood M (2006). Recipes for survival: controlling the bushmeat trade Ape Alliance Report. from Robinson JG, Bennett EL (2000). Hunting for sustainability in tropical forests. New York, Columbia University Press. Robinson JG, Redford KH, Bennett EL (1999). Wildlife harvest in logged tropical forests. Science 285: Rusch GM, Stokke S, Røskaft E, Mwakalebe G, Wiik H, Arnemo JM, Lyamuya RD (2005). Human-wildlife interactions in western Serengeti, Tanzania - effects of land management of migratory routes and mammal populations densities. NINA Report Trondheim, Norwegian Institute of Nature Research 85: 47. Sachedina, HT (2008 ). Wildlife is our oil: Conservation, livelihoods and NGOs in the Tarangire Ecosystem, Tanzania. Thesis submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy. Oxford University Centre for the Environment. Oxford, UK, University of Oxford. D.Phil. Sinclair ARE and P Arcese Serengeti in the context of worldwide conservation efforts. Serengeti II: Dynamics, management and conservation of an ecosystem. ARE Sinclair and P Arcese. Chicago, Ill, University Press of Chicago: Starkey M (2004). Commerce and subsistence: the hunting, sale and consumption of bushmeat in Gabon. Cambridge, UK, Fitzwilliam College, Cambridge University. Thirgood S, Mosser A, Tham S, Hopcraf G, Mwangomo LE, Mlengeya T, Kilewo M, Fryxell J, Sinclair ARE Borner M (2004). Can parks protect migratory ungulates? The case of the Serengeti wildebeest, Anim.. Conserv. 7: URT (1974). The Wildlife Conservation Act, 1974, The Gazette of the United Republic of Tanzania. LV. Nov. URT (2002). Population and housing censuses. Tourism-Department. Dar es Salaam, Tanzania, Ministry of Natural Resources and Tourism. URT (2002). The wildlife conservation (tourist hunting) regulations, 2000: revised edition, Dar es Salaam, Tanzania, The Gazette of the United Republic of Tanzania. Wilkie DS, Carpenter JF (1999;8). Bushmeat hunting in the Congo Basin: an assessment of impacts and options for mitigation, Biodiv. Conserv. pp Wilkie DS, Malcolm S, Kate A, Ernestine NE, Paul T, Ricardo G (2005). "Role of prices and wealth in consumer demand for bushmeat in Gabon, Central Africa." Conserv. Biol. 19: