干旱区生态环境知识资源中心

Numerous territories of Sub-Saharan Africa are composed of a mosaic of very different landscape units: cropland, forest, and savannah. This spatial, but also temporal heterogeneity leads to complexity in the analysis of cattle intake behaviour. The instantaneous intake rate (IIR) is generally analysed in relation to forage biomass density (Bm), i.e. the functional response. We analysed the relationship between IIR and Bm and effects of other vegetation and animal factors for two herds of N’Dama cattle grazing freely during the dry season in the complex Sare Yoro Bana landscape located in the Kolda region in Southern Senegal. The available forages included crop residues (from rice, millet and maize), grass and hay, fruits and litter. The amount and quality of biomass declined throughout the dry season. The variation in IIR was large and could be attributed to differences in Bm (for 30%) and differed between seasons and vegetation types (also 30% of variation explained). Effects of animal characteristics such as wither height and requirements were significant but small compared to herbage-related factors. In the middle of the dry season (MDS) the IIR was higher than in the early dry season (EDS), probably due to better herbage prehensibility, whereas in the late dry season total herbage stocks were very low and also IIR was lower than in EDS and MDS. Two different relations were observed between IIR and the bite rate. In the EDS higher bite rate resulted in higher IIR, compared to an absence of such effect in later stages of the season indicating a compensation between ingestion rate and bite size. Based on these result, we conclude that the relation between intake rate and biomass density generally follows a type-2 functional response, modelled by a monotonically saturating function, but is strongly affected by many factors related to herbage palatability. Despite the inherent spatio-temporal complexity of the vegetation and the challenges in observations and data collection these relationships can be quantified for cattle grazing in a heterogeneous semi-arid landscape. (C) 2014 Elsevier B.V. All rights reserved.