The removal of trees in rangelands can create gaps and lead to increased grass production that could suppress subsequent tree seedling establishment and growth. However, gaps can also enhance the growth of remaining trees. We conducted a field experiment at two savanna sites with different soil texture and woody species. We used 24 plots at each site to determine the effect of tree-removal intensities (0%, 10%, 20%, 50%, 75% and 100%) on grass production, tree-seedling establishment and growth, and growth of the remaining large trees. Site 1 was on previously cultivated severely-eroded clay-dominated soils, encroached by a monospecific stand of Vachellia tortilis. Site 2 had never been cultivated, and was on sandy soils with several woody species. At Site 1, 75 and 100% tree removal significantly reduced standing grass biomass towards the end of the first growing season, with no differences towards the end of the second season. At Site 2, tree removal significantly increased standing grass biomass. There was no significant effect of tree removal on tree seedling establishment at Site 1, but at Site 2 tree removal had a significantly negative effect on overall tree seedling establishment. At both sites, there were no significant differences in tree seedling growth. Moderate (50%) to high (75%) removal of trees had a positive effect on the growth of remaining large trees at both study sites. We found that tree seedling establishment could be affected by the level of grass biomass following tree removal, but other factors including soil erosion are also important considerations. Reduced tree competition facilitates growth of remaining large trees. An implication of these findings is that, regardless of the substantial costs of woody plant control, the recovery of key ecosystem services such as an increased forage production may not be realised. However, we recognise that this may be system-specific.

Mechanical and chemical methods are widely used to control woody plant encroachment in many African countries. However, very little is known about the effectiveness of these control methods among woody species of different ages. We conducted a field experiment to determine the effects of different tree removal treatments (10%, 20%, 50%, 75% and 100%) and herbicide application (Picloram; 6 mL L⁻¹) on the resprouting ability and vigour of 12 woody plant species. We examined 20 plots (30 m × 30 m) that were each subjected to tree removal, followed by herbicide application on half of the stems for each plot. All the tree species in this study resprouted after cutting. The applied concentration of herbicide significantly reduced the shoot production for Ehretia rigida, Vachellia robusta and Ziziphus mucronata, with a marginal effect for Dichrostachys cinerea. The diameter of stems was an important factor in determining resprouting ability, with shoot production decreasing with increasing stem diameter. However, stem diameter did not affect shoot length and diameter for all species. We found that woody plants are more likely to resprout and survive as juveniles than as adults after cutting and that herbicide only affected four of the twelve species at a concentration of 6 mL L⁻¹. Thus, testing the amount of Picloram needed to kill certain woody species may be of importance for land users in southern African savannas.

Along with the woody plant expansion that is predicted to continue at the expense of the grassy layer is the increasing societal demand for animal protein and livestock products. Unless concerted efforts by land users, ecologists, and animal scientists are made to increase the utilization of trees and shrubs as forage, it will be impossible to meet future demand for meat and meat products. We conducted two short-term pen experiments to determine the effects of (1) supplementation with polyethylene glycol (PEG-a polymer purported to bind and neutralize the negative effects of tannins), a high-protein source (soybean meal), and a high-energy source (yellow maize grain) and (2) diet mixing (single-species vs. multispecies diets) on the intake of condensed tannin-rich woody plants (i.e., Searsia lancea, S. pyroides, and Euclea crispa) by goats. While all three forage species were used in the diet mixing experiment (Exp. 2), only E. crispa was used in the supplementation experiment (Exp. 1). Supplementing goats with energy- and protein-rich sources significantly increased the intake of E. crispa (p < 0.05), 713.4 g ± 13.5 and 760 g ± 28.9, respectively, whereas those on the control diet maintained their intake at 540.32 g ± 11.2. Although PEG tended to increase the consumption of E. crispa by goats, the observed increase was not significant (p > 0.05) from that observed in other treatments. In the diet mixing experiment, goats offered a combination of all three forage species attained substantially higher dry matter intakes compared to the goats offered these species individually (p < 0.05). While longer-term field experiments are needed in the African savannas, we postulate from the current results that management strategies that provide animals with (1) a variety of species in the diet vs. monocultures and (2) a combination of nutrient-rich and tannin-rich species may improve the ability of goats to consume chemically defended woody plants.