Change in land use and land cover (LULC) is an area of interest in landscape ecology due to its adverse environmental impacts on wildlife habitats. At local, regional, and global scales, the fragmentation and obstruction of wildlife corridors has considerably increased over time as a result of factors connected to land use and climate variables. These environmental changes are a threat to wildlife and biodiversity conservation, particularly in places where large iconic mammals are supposed to roam freely. This study was carried out in North-Western Botswana, in the Chobe enclave, a communally managed area. The enclave is surrounded in the west by the Chobe National Park, the Namibian border (Cuando-Linyanti River) to the north and east and the Chobe Forest Reserve in the south. Seasonal flooding of the Cuando-Linyanti River system creates regular transitions between mesic and the local semi-arid climate, making this floodplain a haven for wildlife including the big games. Botswana has the highest freely roaming African savannah elephants’ population in Africa, estimated at 126 000 in 2018 and largest concentrations are found in the northern part of the country. There is dearth of knowledge about the impacts of landscape fragmentation on human elephant interactions in Botswana Particularly in Chobe enclave. Chobe enclave is experiencing increasing human population and large concentrations of freely roaming elephants. This have created competition between humans and elephants for space and other resources. It is vital to comprehend how landscape fragmentation and processes influences elephant’s movements and ultimately human-elephants conflict especially in ecologically valued areas such as the Chobe enclave. With knowledge on how LULC dynamics, fragmentation and elephant movements and vii distribution influences one another, possible and informed measures can be proposed to abate adverse human elephant interaction. The aim of the study was to characterize and quantify LULC change using Landsat time series and to assess the influence of landscape fragmentation on human-elephant interaction (HEI) in the Chobe enclave over a period of 20 years from 2000 until 2020 using landscape metrics and field evidence information. Multi-temporal satellite imageries for all the study periods were ordered and downloaded from the USGS webpage at Landsat scene 174/072 and 174/073. Landsat imageries were radiometrically and geometrically corrected for best classification results. A classification scheme consisting of five LULC classes was established. The maximum likelihood classifier was employed for supervised classification. Classified maps accuracy was measured and change detection were performed on classified maps. In accordance with the purpose of this study, five (5) indices were selected for spatial characterization of the study area at class and landscape level. Selection of which landscape metrics to use was based on their ability to act as indicators of landscape change and also on their ability to quantify and assess landscape fragmentation. Selected indices, class area (CA), number of patches (NP), largest patch density (LPI), landscape splitting index (LSI) and aggregation (AI) were computed in a software called FRASTATS. Semi-structured interviews were administered to a total of 57 purposely selected key informants to seek knowledge and validation on LULC and HEI. High resolution images in Google Earth Pro together with indigenous knowledge were used to identify and delineate elephant migratory routes in the study area. Human elephant conflict hotspots (HECH) for the study period 2000, 2020 and indigenous knowledge were derived from the kernel density estimation. Results from LULC analysis shows that in 2000 the study area was dominated by Shrubland, 70 383 ha (45.2%) while in 2010 and 2020 LULC dominance shifted to grassland, 44 270 ha (29.5%) and 56 935 ha (31.2%) respectively. Change in the share of land in the study area is attributed to land use intensification, wild fires, over browsing and vegetation destruction by elephants and droughts. Landscape fragmentation within elephant habitats and migratory corridors is a result of land use encroachment thus creating land use conflicts scenarios between local communities and elephants. Landscape metrics reveals that the study period 2020 has the most fragmented landscapes with high cases of HEC. The fragmented landscape is characterized by subdivisions or landscape patches, low aggregation index and high landscape splitting index. Human elephant conflict hotspots positively correlate with landscape fragmentation. Socio economic impacts associated with HEIs in the study area includes negative attitude towards elephants by the enclave residents, indirect financial loss due to elephant crop raiding and property destruction. The study proposes that wildlife migratory routes in the Chobe enclave be marked and incorporated in land use plans and zones to avoid further habitat fragmentation and expansion of land use into them. This could enhance peaceful coexistence between people and elephants as conflicts would be minimized.

Thesis (MSc in Environmental Science)---Botswana International University of Science and Technology, 2023