Effects of Wildlife-livestock-human Interactions on habitat in the Meru National Park and Bisanadi National Reserve
The study was designed to characterise the linkages between wildlife-livestock- human conflicts and resource access in the Meru Conservation Area (MCA), determine the impact of changes in land use on land cover between 1987 and 2001, and describe the vegetation structure and composition as a baseline for future assessment of the impact of present conflict mitigation strategies. A questionnaire survey was conducted among 80 households randomly distributed in the western, southern and northern buffer zones of the MCA. Human population dynamics between 1980 and 2000, land use patterns, wildlife-livestock-human conflicts and their causes, and conflict mitigation strategies were tabulated for analysis using descriptive statistics. The z-t e st for paired proportions was used to determine the significance of change In human population. The Chi-square test for the goodness of fit was used to determine the level of household exposure to conflicts, while correlation analysis was used to evaluate the relationship between the level of household exposure to wildlife-livestock-human conflicts and distance from the protected area boundary. Habitat changes were determined through geographic information systems (GIS) and remote sensing analysis using Landsat Thematic Mapper images of 1987 and 2001. The observed differences were tested for significance us in g-. the z-test of one sample set. Vegetation description involved a hierarchical classification based on physiognomic and floristic attributes. Plant life-form, stratification, composition and cover were some of the features considered. The paired r-test indicated a significant increase in human population (tI58; P<0.05) in the buffer zones surrounding the MCA between 1980 and 2000. The survey recorded "an 83% increase in human population in the MCA's western, southern and northern buffer zones, a figure that compared closely to the 86% increase in human p o p u l a ti o n reported in. the 1999 household census by the Central Bureau of Statistics in the six districts bordering the MCA, which indicated that human population increased from 1.56 million in 1979 to 2.90 million in 1999. Thus, the human population density in the MCA increased from an average of 2-50 'people per km2 to 51-125 people per km2 in the southern Tharaka zone, and an average of 126-500 people km2 to over 500 people per km2 in the western buffer zones bordering Nyambene Ranges. The survey demonstrated that 70% of the households occupying the buffer zones surrounding the MCA were migrant communities who came to the area between 1980 and 2000. This implies that the observed increase in human population was largely attributable to immigration rather than population Increase among the pre-existing population. Analysis of migrant households by land use indicated that 42.9% and 44.6% were agricultural and agropastoral communities respectively. Majority of the agricultural households migrated from nearby high potential agro-ecological districts, such as Meru South, Nyeri and Murang'a in Central Province. About 96% of agro-pastoral households migrated from neighbouring Isiolo, Tharaka, Meru South, Meru Central and Meru North Districts. Pure pastoralists, who constituted 12.5% of the households that settled in the MCA between 1980 and 2000, migrated from Isiolo, Garrisa and Mandera In northern Kenya. This led to the introduction of cropping agriculture In the buffer zones surrounding the conservation area and the subsequent fragmentation and alteration of habitats that formerly served as communal grazing lands and wildlife dispersal areas, particularly on the western and southern boundaries of Meru National Park. The situation led to a reduction of the livestock grazing range and the confinement of most wildlife to the protected area. Consequently, the livestock/wildlife resource base began to decline leading to increased resource competition as wildlife; livestock and agriculturalists sought access to limiting resources, a situation which led to increased wildlife-livestock-human interactions. The most notable indicators of these interactions were wildlife-livestock-human conflicts and habitat changes In the western, southern and northern buffer zones of the MCA. Three principal land use types were identified In the MCA, namely: agriculture, agro-pastoralism and pure-pastoralism. Agriculture was practised by 35% of the households interviewed, majority of which occupied the relatively wetter western buffer zones of the conservation area (with about 1000mm of rainfall year-I) that border the high potential areas of Nyambene Ranges. Agro-pastoralism was: practised by 55% of the households interviewed. Although most of the agro-pastoral households occupied the southern Tharaka zone, an appreciable number (20%) 'were encountered in the northern pastoral lands that border Bisanadi National Reserve to the north. The agro-pastor al households in the northern pastoral lands comprised mainly of pure pastoralists who had settled down over the past decade to practice crop cultivation. The situation is likely to have arise n from extensive conversion of pastoral grazing land into farmland especially in the MCA's western and southern buffer zones, which may have led to the sedentarization of pure pastoralists and hence the need to spread risks through diversified land use. About 10% of the households practised pure-pastoralism. These households consisted mainly of the Borana pastoral community who occupied the northern pastoral lands, but owing to lack of formal boundaries in their grazing patterns, they could be spotted in the communal grazing lands and wildlife dispersal areas of the western Meru National Park boundary towards the southern Tharaka areas, particularly in the dry season. Wildlife-livestock-human conflict types in the MCA were grouped into two broad categories: conflicts affecting communities bordering the MCA; and conflicts affecting wildlife. Conflicts affecting communities bordering the MCA included crop destruction by wildlife, livestock attacks by wildlife, human attacks by wildlife and disease transmission from wildlife to livestock. Over 90% of the households interviewed were exposed to these conflicts leading to a very significant difference (X2; df=2; P< 0.01) in the Chi-square test for the goodness of fit between the expected and observed levels of wildlife-livestock-human conflicts. However, the level of exposure to these conflicts was found to be dependent on the distance of homestead / land use activity from the protected area boundary, the type of conflict, land use type, and strategies used by different households to mitigate conflicts. On the other hand, correlation analysis of the duration of stay in the MCA and the distance of homesteads and/or land use from the protected area boundary indicated a very strong link (P < 0.01) between the level of exposure to wildlife-livestock- human conflicts and the distance from the protected area boundary. This implied that households that migrated to the MCA between 1980 and 2000 tended to stay closer to the protected area boundary than those households that occupied the. MCA before 1980. For instance, the average distance from the park boundary for the 72.5% of households that were highly exposed to conflicts was 2.6kl1). and· they had stayed in the MCA for an average of 16 years. Households which were moderately exposed to conflicts were 16.3% and they had stayed in the MCA for an average of 31 years at about 6.8km away from the park boundary. Households experiencing negligible exposure to conflicts were 11.2% and they had stayed in the MCA for an average of 32 years at about 8.5km away from the park boundary. Conflicts affecting wildlife included the encroachment of pastoral livestock into the protected area for forage and water in the dry season, the fragmentation of areas surrounding the protected area into human settlements and farmlands by migrant agriculturalists, and poaching of wildlife. Wildlife-livestock-human conflicts in the MCA were attributed to four main causes, namely: causes related to land use change and Kenya's ASAL land use policy; causes associated with community perception and attitudes towards the protected area, causes attributable to wildlife conservation and utilization policy; and those associated with KWS' ineffectiveness In wildlife management. Conflicts related to land use change were attributed to present ASAL land use policy in Kenya, which in an effort to address rapid increase in human population and the need for self-sufficiency in food production, tends to promote dry land agriculture in ASAL areas at the expense of pastoral livestock production and wildlife management. Under this policy regime, agricultural households have continued to encroach into the MCA's wetter margins, such as the western and southern buffer zones, where they have fragmented and altered dry season Iivestock grazing reserves and wildlife dispersal areas into human settlements and smallholder cultivation units. Thus, the livestock/wildlife resource base has continued •. to diminish leading to increased wildlife-livestock-human interactions. Conflicts associated with com m unity perception and attitudes towards the protected area arose from the pastoral notion that the park has taken all the best areas, with regard to pasture and water. They often failed to realise that even areas available to them in the buffer zones surrounding the MCA would look as good if subjected to proper range c cnservation principles. They questioned the rationale behind setting aside large tracts of land as protected areas when the aim is only to keep wildlife for 'people to view', an argument that is understandable from the pastoral (grazing) perspective, but not from a conservation perspective. The negative pastoral perception of the MCA arose, partly, from their herd stocking rates. The MCA, for instance, recorded a 97.7% increase in the livestock population among pure pastoral communities occupying areas north of Bisanadi National Reserve between 1980 and 2000 . However, most the pastoralists kept the local Zebu breed, which has both slow maturity and low productivity rates. The result was that livestock numbers became more important that livestock quality, which eventually led to resource (forage and water) conflicts between livestock and wildlife, particularly in the dry season. Conflicts attributable to wildlife conservation and utilization policy arose mainly from contentions regarding the existing legislative policy, which prohibits the consumptive utilization of wildlife resources even when communities living with wildlife continue to suffer the destruction of property and loss of human lives from attacks by wildlife outside the formal boundaries of protected areas. Thus, the question that featured prominently among communities living with wildlife in the MCA was: who bears the burden of wildlife conservation; and who benefits? Under these conditions, it has increasingly become difficult to commit local communities to community based wildlife conservation, thereby leading to the continued encroachment of cropping agriculture into wildlife dispersal areas in the MCA. Conflicts associated with KWS' ineffectiveness in wildlife management were attributed to the organization's law enforcement approach to conflict management as opposed to consensus building with communities living with wildlife. KWS was observed to be reactive in its approach to wildlife-livestock- human conflicts, a situation which created the impression that wildlife belonged to the government, even when local communities considered the government to be in office by the people's mandate. One of the intervention measures employed by KWS to manage wildlife-livestock- human conflicts in the MCA was to fence off Meru National Park and Bisanadi National Reserve. However, given the long-term ecological implications of this initiat iv,e for biodiversity - conservation, this study described the vegetation in the enclosed area to create an inventory for future assessment of the eff'ectiveness of .fencing as a human-wildlife conflict management strategy. Six physiognomic classes with various floristic attributes were described, namely: Acacia-Chloris & 'Combr etum-Sehim a open wooden grassland; Acacia-Combretum-Chloris shrub grassland; Acacia- Commiphora bushland vegetation; Phoenix-Hyphaene-Raphia-Sporobolus riverine vegetation; Hyphaene-Chlor is impeded drainage vegetation and Xerophyta-Euphorbia-Albizia inselberg vegetation. Remote sensing analysis indicated significant land cover changes in the MCA between 1987 and 2001, particularly in the western, southern and northern buffer zones. The area under cultivation in the western and southern buffer zones increased by 76% - from 35,054 ha in 1987 to 61,853 ha in 2001. On the other hand, the northern pastoral lands experienced bush and shrub encroachment into the shrub grassland vegetation of 1987 as a reducing grazing range in the western buffer zones compelled pastoralists to subject the area to pro longed heavy grazing. In the southern Tharaka area, the dense bushland vegetation of 1987 had become sparse as shifting cultivation extended into it, but remained relatively dense inside the park. Overall, increased wildlife-livestock-human interactions in the MCA and their indicators were both directly and indirectly traced to land use change and a diminishing resource base. The underlying causes were identified to be present policies on ASAL land use, wildlife conservation and utilization, and human population growth and distribution. The study recommended that if livestock and wildlife habitats in Kenyan rangelands are to be conserved, integrated policy formulation may be necessary to safeguard traditional ASAL land use types, such as pastoral livestock production and wildlife management.
CitationMaster of Science in Plant Ecology
University of NairobiDepartment of Botany