Seasonal Declines in Rangeland Quality in the Cultivated Hardveld of Botswana Using Landsat Thematic Mapper Imagery

Abstract

The need to combat both seasonal and long‐term drought are endemic in semi‐arid countries such as Botswana. These problems entail inter alia locating browse and certain bush derived foodstuffs (veld products) year long to enable local people to sustain their livelihoods. Seasonal range condition data are also required to help minimise overstocking and degradation. Results from seasonal reflectance curve analysis from southeastern Botswana indicate that most vegetation types can be identified especially on longer Thematic Mapper (TM) wavebands. Spectral Separability Indexes (SSIs) confirmed that range feature separation varies as a function of waveband and that feature separation is more pronounced in the wet, than the dry‐season. Results show that range feature differentiation is best attained using a combination of the chlorophyll absorpance band (TM3) and two mid‐infrared bands (TM5 and TM7). In terms of total seasonal separability, wet‐season vegetation cover differentiation was found to be 1.7 times more effective than that derived from dry‐season data. Both wet and dry‐season data sets were classified separately using maximum likelihood techniques. Although wet‐season data were more accurately classified, satisfactory class coincidence was achieved across the two data sets. Results indicate dry‐season declines in high quality rangeland and increases in bush encroached areas especially in communal areas comprising scattered villages. An appraisal of the ability of rangeland to sustain the needs of the rural poor in semi‐arid areas needs realistically to be undertaken during the dry season. Even though this time is sub‐optimal in terms of range condition assessment, it corresponds to the winter months when the demand for fuelwood is at its peak and available forage is minimal. Optimal timing, high resolution imagery and knowledge of local conditions are required to develop an information base which can help inform policy and direct poverty alleviation strategies.