Smit, G.N.Malan, P. J.Boys, Jerome Marcelino2023-10-132023-10-132022http://hdl.handle.net/11660/12301Thesis (Ph.D.(Grassland Science))--University of the Free State, 2022The savannas of northern Namibia are prone to woody plant thickening, and land owners, therefore, attempt to restore rangelands through the removal of some or all of the woody plants with a view to reduce the negative competition interactions with the herbaceous layer, allowing an increase in the grazing capacity for livestock. The utilization of the bush resource made available through these bush control programs has led to the development of a lucrative bush value-chain industry which often leads to the over-exploitation of the resource for short-term financial gains. The effect of the most common harvesting practices such as non-selective and selective woody biomass harvesting was investigated over a three-year period in an area northwest of Otjiwarongo, described as the North-central plateau of Namibia and with vegetation classified as Thornbush Savanna. The dominant woody plant species were Senegalia mellifera and Vachellia reficiens. They are also the most sought-after species for charcoal production for their high wood yield potential, and for this reason they are also the species of interest in this study. The coppicing ability, survival, change in woody species composition, woody plant recruitment through coppice and/or sapling establishment, rate of biomass production, effect of annual total coppice defoliation on coppice regrowth survival of selected species, and growth rates and mortalities of woody plants were studied. The trail layout consisted of plots of 50 x 30 m (1 500m²) with at least two replications of each treatment. The non-selective (bush-roller and chainsaw) harvesting showed no significant differences (P>0.05) in coppice regrowth of Senegalia mellifera and Vachellia reficiens. However, although not significant (P>0.05), the chainsaw harvested plots had higher coppice rates when considering all other woody species whereas, bush-roller harvested plots had a higher sapling establishment as compared to the chainsaw harvested plots. Dichrostachys cinerea was the only species that had a highly significant (P=0.002) increase in sapling establishment. There was a change in dominance from high wood-potential (mostly S. mellifera, V. reficiens, Vachellia tortilis, V. hebeclada) to low wood-potential (mostly all Grewia species, D. cinerea, Catophractes alexandrii) species. The ETTE ha⁻¹ is projected to return to its original pre-harvest state in less than 12 years, whereas the wood biomass did not show any signs of returning to its original pre-harvest state. Annual total defoliation of coppice proved to be effective for the control of coppicing woody plants without the need for any chemical treatment (arboricide). There was a highly significant difference (P<0.01) in the rate of coppice between the two seasons (hot wet summer and cold dry winter) as treatments. Plants initially harvested in summer had a higher coppice and survival rate than plants harvested in winter. The summer harvested S. mellifera stumps coppiced more vigorously compared to V. reficiens. The winter harvested V. reficiens coppiced more vigorously compared to S. mellifera. There were highly significant differences (P<0.001) in coppice amongst the selectively versus non-selectively harvested plots. Plants harvested selectively, while retaining 4 500 ETTE ha⁻¹, had lower coppicing rates than those where the whole area was harvested non-selectively. The high ETTE ha⁻¹ before harvesting could not be matched three years after harvesting, whereas plants ha⁻¹ in the selectively thinned plots significantly (P<0.05) exceeded the original plant density (plants ha⁻¹) before initial harvesting within three years. Mature S. mellifera woody plants had a significantly higher (P<0.001) mortality rate in the bush-thickened control plots. The mortality of V. reficiens was not as high as that of S. mellifera. Regarding shoot growth, the growth in shoot length was more pronounced than that of shoot diameter. Vachellia reficiens had a significantly higher (P=0.001) growth rate in shoot length compared to S. mellifera. Allometric regression equations were also developed from harvested undamaged Senegalia mellifera and Vachellia reficiens plants for inclusion into the BECVOL-3 model. Highly significant relationships (P<0.01) were found between the spatial canopy volume and the dependent biomass variables predicted. There were negligible differences in percentage species composition of herbaceous plants amongst the different bush control treatment plots. Although surveys in all plots confirmed a state of bush thickening as per the general rule of thumb and thus the need for some form of bush thinning, the grass dry matter production did not improve, as generally assumed, with reduced woody plant competition. The herbaceous dry matter production showed a more pronounced positive response to increased rainfall rather than bush thinning during the study period. The first year of the study had a below-average annual rainfall and the rest of the years followed with above average annual rainfall.enBush-thickeningnon-selective harvestingselective thinningcoppicesapling establishmenthigh wood-potential specieslow wood-potential speciesprotected speciesrestorationThesisUniversity of the Free State