Relations between woody and herbaceous components and the effects of bush‐clearing in southern African savannas

Publisher Website
Google Scholar
Abstract
The relations between woody and herbaceous plants in southern African savannas are reviewed in the context of pastoral use. Current knowledge is summarized, important gaps are indicated and research priorities are suggested. Although all savannas consist of a grass and a woody component, the physical determinants, biological interactions and individual species properties are unique to each spatial and temporal situation. In addition, past management has resulted in different kinds and degrees of modification. Primary and secondary determinants are discussed in terms of how they influence the structure and productivity of savanna communities. More than one outcome is possible to any competitive interaction owing to the modification of the effects of determinants by changing weather sequences, asymmetrical competition and biological interactions. This contrasts sharply with the conventional rangeland succession model. Relations between woody and herbaceous plants are often discontinuous because of the dynamic interactions between determinants, plants and herbivory. There is an apparent discrepancy between the convex and concave response of grass productivity to increasing woody biomass. It is hypothesized that a single discontinuous function explains this apparent discrepancy. This unifying theorem accounts for the modifications due to changes in levels of determinants and different species mixes within and between sites. With a few notable exceptions, research in southern Africa has been of an ad hoc, empirical and limited nature. It has not aimed at achieving an understanding and a predictive capability beyond the site and circumstances of individual trials themselves. A case is made for a systems‐orientated research approach, so that site and situation‐specific problems can be solved to fulfil management objectives.

This publication has 46 references indexed in Scilit: