Disturbances, such as herbivory and fire are commonplace in savanna ecosystems. The effects of herbivory and fire on growth and defences of adult trees is a much studied filed in plant ecology. However, there are comparatively few studies that have investigated the effects of herbivory and fire on seedling growth, defences and establishment. African Acacia trees are common and widespread, and are suggested to be keystone species in savanna ecosystems. They have been shown to significantly positively increase soil characteristics, such as soil moisture and infiltration, while also increasing spatial heterogeneity of savannas. These trees are distinguishable by their array of physical and chemical defences. Physical defences are either in the form of spines (physiologically costly to produce, thus considered inducible) or prickles (physiologically cheap to produce, thus considered constitutive), while condensed tannins are the most common form of chemical defences. Adult Acacia trees have been shown on several occasions to be highly resilient to disturbance events, primarily due to their large size. However, the effects of herbivory and fire on Acacia seedlings have been little studied despite their apparent importance for our understanding of African savanna ecosystem functioning. In two separate experiments, this thesis aimed to investigate the individual and combined effects of simulated herbivory and fire on the regrowth, defences, and total non-structural carbohydrates (TNC) of the seedlings of several Acacia species, while also testing the Resource Availability Hypothesis (RAH) (14 Acacia species) and the Expanded Growth-Differentiation Balance Hypothesis (GDBH) (three Acacia species). The RAH (Coley, Bryant & Chapin 1985) and the GDBH (Loomis 1932) both assume that arid-adapted (resource-poor) species are slow growing, with low tissue turnover rates. Contrastingly, humid-adapted (resource-rich) species have fast growth rates and thus a high tissue turnover rate. Therefore, arid-adapted species are predicted to invest more carbohydrate reserves in defence after a disturbance event, in order to defend new photosynthetic material. Conversely, humid-adapted (resource-rich) species are predicted to invest carbohydrate reserves into increased growth after a disturbance event, in order to compensate for tissue loss. The first greenhouse experiment found that, in accordance with the RAH, humid-adapted species displayed elevated growth rates compared to arid-adapted species. Overall, defences significantly increased after herbivory, but significantly decreased after fire. Herbivory was also shown to significantly reduce TNC stores in roots more than fire. We found that arid-adapted species did indeed invest more carbohydrate stores (TNC) into defence, and displaying an increase in spine and prickle abundance, spine and prickle length, and condensed tannin concentration. Humid-adapted species displayed an increase in growth rather than in defence, with the majority of species only increasing one defensive trait. Humid-adapted species also displayed significant trade-offs (negative correlations) between new stem growth and defence traits, while arid-adapted species overall did not display any significant trade-offs between stem growth and increased defences. The majority of arid-adapted species also displayed an allometric effect for spine abundance and length, with an increase in one trait led to an increase in the other. Prickles were found to be inducible, despite the assumption that due to low physiological cost, they are no inducible. Herbivory and fire were shown to not be substitutable in their effects on Acacia seedlings in a controlled experimental setting. The second greenhouse experiment tested the Expanded Growth-Differentiation Balance Hypothesis (GDBH) in the seedlings of A. erioloba, A. karroo, and A. nilotica using five levels of nutrient availability. Overall, spine abundance and spine length displayed a unimodal trend in all three species, with spine abundance and spine length being greatest at a nutrient availability of 800 mg/ ℓ. Spine abundance, spine length and condensed tannins increased significantly after herbivory, but were shown to significantly decrease after fire. We found that with an increase in nutrient availability, the growth of stems and roots, along with the production of TNC in roots and stems of all species significantly increased. Our data provide mixed support for the assumptions and predictions of the expanded GDBH. The regrowth of stems and physical defences were consistent with the GDBH. Chemical defences (i.e. condensed tannins) were however, inconsistent with the predictions of the GDBH. We have shown that Acacia seedlings are highly resilient to disturbance events, while the growth responses of Acacia seedlings are underpinned by TNC stores in roots. Simulated herbivory and fire are often substituted for one another in controlled experiments. However, we have shown that the effects of herbivory and fire have significantly different effects on regrowth patterns of Acacia seedlings. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2011.

A major challenge in savanna rangeland studies is estimating woody vegetation cover and densities over large areas where field based census alone is impractical. It is therefore crucial that the management and conservation oriented research in savannas identify data sources that provides quick, timely and economical means to obtain information on vegetation cover. Satellite remote sensing can provide such information. Remote sensing investigations, however, require establishing statistical relationships between field and remotely sensed data. Usually regression is the empirical method applied to field and remotely sensed data for the spatial estimation of woody vegetation variables. Geostatistical techniques, which take spatial autocorrelation of variables into consideration, have rarely been used for this purpose. We investigated the possibility of improving woody biomass predictions in tropical savannas using cokriging. Cokriging was used to evaluate the cross-correlated information between SPOT (Satellites Pour l’Observation de la Terre or Earth-observing Satellites)-derived vegetation variables and field sampled woody vegetation percentage canopy cover and density. The main focus was to estimate woody density and map the distribution of woody cover in an African savanna environment. In order to select the best SPOT-derived vegetation variable that best correlate with field sampled woody variables, several spectral vegetation and texture indices were evaluated. Next, variogram models were developed: one for woody canopy cover and density, one for the best SPOT-derived vegetation variable, and a crossvariogram between woody variables and best SPOT-derived data. These variograms were then used in cokriging to estimate woody density and map its spatial distribution. Results obtained indicate that through cokriging, the estimation accuracy can be improved compared to ordinary kriging and stepwise linear regression. Cokriging therefore provided a method to combine field and remotely sensed data to accurately estimate woody cover variables. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2008.

Thesis (MScFor)--Stellenbosch University, 2002. / ENGLISH ABSTRACT: Increases in woody plant density ('bush encroachment') reduce livestock production and biodiversity. By convention, soil moisture, soil nutrients, fire and herbivory are regarded as the principal factors governing the tree-grass ratio of savannas. An experiment with a completely-crossed design was employed to investigate woody seedling (Acacia me/lifera) recruitment near Kimberley, Northern Cape, South Africa, after fire and under conditions of maximum-recorded rainfall, nitrogen addition and grazing. The field experiment was repeated as a garden experiment to determine if the two experiments yield comparable results. Tree germination in the field was extremely low, probably due to below-average natural rainfall in plots that only received natural rain, and insufficient watering frequency in irrigated plots. As a result of low germination, none of the treatments (rain/nitrogen/fire/grazing) had a significant effect on tree recruitment in the field experiment. The duration of the experiment (2000/2001 growing season) was insufficient for the treatments to affect grass composition, although the high rainfall treatment and grazing exclusion significantly improved grass cover and height. The garden experiment showed that frequent watering, no nitrogen addition and grass clipping significantly enhanced tree germination and survival (termed 'recruitment'). There were also significant interactions among rain, nitrogen and grazing in their effects on tree recruitment. The effects of rain on tree recruitment were more pronounced under nitrogen supplementation and vice versa. Similarly, high rain and high nitrogen enhanced the effect of grazing on tree recruitment. It is inferred that above-average rainfall years with frequent rainfall events are required for mass tree recruitment. Tree seedlings can further benefit from space and resources which are made available through grass defoliation. Conversely, nitrogen enrichment improves the competitive ability of the fast-growing grasses relatively more than that of the N2-fixing tree component, thereby suppressing tree recruitment. In contrast to conventional wisdom that grazing alone causes encroachment, it is suggested that there are complex interactions between the abovementioned factors and 'triggering' events such as unusually high rainfall. Contrary to many claims that equilibrium models are inappropriate for explaining savanna dynamics, it was shown that consumer-resource theory has explanatory power for bush-grass dynamics of the savanna studied. The state-space approach that was used facilitated the understanding of savanna dynamics and enabled predictions about the system's response to perturbations. The applicability of consumer-resource theory to semi-arid nutrient-poor savannas confirmed the importance of resource competition in structuring natural systems. / AFRIKAANSE OPSOMMING: Toenames in die digtheid van houtagtige plante ('bosverdigting') verlaag veeproduksie en biodiversiteit. Volgens konvensie word grondvog, grondvoedingstowwe, vuur en beweiding beskou as die belangrikste faktore wat die boom-gras verhouding van savannas bepaal. 'n Volledig-gekruisde ("completely crossed") eksperiment is ontwerp om boomsaad-ontkieming (Acacia mellifera) naby Kimberley, Noord-Kaap, Suid- Afrika, te bestudeer na 'n brand en onder toestande van maksimum-aangetekende reënval, stikstof toevoeging en beweiding. Die veldeskperiment is herhaal as 'n tuineksperiment om vas te stel hoe resultate van die twee eksperimente vergelyk. Boomsaad-ontkieming was uiters laag in die veld, waarskynlik weens ondergemiddelde reënval in persele wat slegs natuurlike reënvalontvang het, en 'n te lae benattingsfrekwensie in die besproeide persele. As gevolg van lae ontkieming in die veldeksperiment, het geen behandeling (reënval/stikstof/brand/beweiding) boomsaad-ontkieming beduidend geaffekteer nie. Die duur van die eksperiment (2000/2001-groeiseisoen) was te kort vir die behandelings om grassamestelling te beïnvloed, alhoewel besproeiing en geen beweiding die grasbedekking en -hoogte betekenisvol verhoog het. Die tuineksperiment het getoon dat boomsaadontkieming en vroeë oorlewing (genoem 'vestiging') betekenisvol verhoog is deur gereëlde benatting, geen stikstof toevoeging en die sny van gras. Daar was ook beduidende interaksies tussen reën, stikstof en beweiding in hul invloed op boomvestiging. Die reënbehandeling het 'n groter uitwerking op boomvestiging gehad onder stikstoftoevoeging en vice versa. Hoë reënval en stikstoftoevoeging het ook die invloed van beweiding op boomvestiging versterk. Die afleiding is dat bogemiddelde reënvaljare met gereëlde reënbuie 'n vereiste is vir grootskaalse boomvestiging. Boomsaailinge kan verder voordeel trek uit die spasie en hulpbronne wat beskikbaar raak wanneer gras ontblaar word. Daarteenoor verhoog stikstofverryking die mededingendheid van die vinniggroeiende grasse meer as dié van die N2-bindende boomkomponent, met die gevolg dat boomvestiging onderdruk word. In kontras met die konvensionele veronderstelling dat beweiding opsigself bosverdigting veroorsaak, word voorgestel dat ingewikkelde interaksies plaasvind tussen die bogenoemde faktore en ander 'sneller-gebeurtenisse', soos buitengewoon hoë reënval. In teenstelling met baie aansprake dat ekwilibrium modelle ontoepaslik is om savanna-dinamika te verklaar, is getoon dat die verbruikers-hulpbron teorie ("consumer-resource theory") oor verklaringsvermoë beskik vir boom-gras dinamika van die savanna wat bestudeer is. Die staat-spasie ("state-space") benadering gebruik, het begrip van savanna-dinamika bevorder en voorspellings moontlik gemaak aangaande die sisteem se reaksie op versteurings. Die toepaslikheid van verbruikers-hulpbron teorie vir semi-ariede, nutriënt-arm savannas bevestig dat kompetisie vir hulpbronne 'n sentrale rol speel in die strukturering van ekologiese sisteme.

Savannahs regulate an agro-ecosystem crucial for the production of domestic livestock, one of the main sources of income worldwide as well as in South African rural communities. Nevertheless, globally these ecosystem functions are threatened by intense human exploitation, inappropriate land use and environmental changes. Leaf area index (LAI) defined as one half the total green leaf area per unit ground surface area, is an inventory of the plant green leaves that defines the actual size of the interface between the vegetation and the atmosphere. Thus, LAI spatial data could serve as an indicator of rangeland productivity. Consequently, the accurate and rapid estimation of LAI is a key requirement for farmers and policy makers to devise sustainable management strategies for rangeland resources. In this study, the main focus was to assess the utility and the accuracy of the PROSAILH radiative transfer model (RTM) to estimate LAI in the South African rangeland on the recently launched Landsat 8 sensor data. The Landsat 8 sensor has been a promising sensor for estimating grassland LAI as compared to its predecessors Landsat 5 to 7 sensors because of its increased radiometric resolution. For this purpose, two PROSAIL inversion methods and semi- empirical methods such as Normalized difference vegetation index (NDVI) were utilized to estimate LAI. The results showed that physically based approaches surpassed empirical approach with highest accuracy yielded by artificial neural network (ANN) inversion approach (RMSE=0.138), in contrast to the Look-Up Table (LUT) approach (RMSE=0.265). In conclusion, the results of this study proved that PROSAIL RTM approach on Landsat 8 data could be utilized to accurately estimate LAI at regional scale which could aid in rapid assessment and monitoring of the rangeland resources. / Environmental Sciences / M. Sc. (Environmental Science)

Savannahs regulate an agro-ecosystem crucial for the production of domestic livestock, one of the main sources of income worldwide as well as in South African rural communities. Nevertheless, globally these ecosystem functions are threatened by intense human exploitation, inappropriate land use and environmental changes. Leaf area index (LAI) defined as one half the total green leaf area per unit ground surface area, is an inventory of the plant green leaves that defines the actual size of the interface between the vegetation and the atmosphere. Thus, LAI spatial data could serve as an indicator of rangeland productivity. Consequently, the accurate and rapid estimation of LAI is a key requirement for farmers and policy makers to devise sustainable management strategies for rangeland resources. In this study, the main focus was to assess the utility and the accuracy of the PROSAILH radiative transfer model (RTM) to estimate LAI in the South African rangeland on the recently launched Landsat 8 sensor data. The Landsat 8 sensor has been a promising sensor for estimating grassland LAI as compared to its predecessors Landsat 5 to 7 sensors because of its increased radiometric resolution. For this purpose, two PROSAIL inversion methods and semi- empirical methods such as Normalized difference vegetation index (NDVI) were utilized to estimate LAI. The results showed that physically based approaches surpassed empirical approach with highest accuracy yielded by artificial neural network (ANN) inversion approach (RMSE=0.138), in contrast to the Look-Up Table (LUT) approach (RMSE=0.265). In conclusion, the results of this study proved that PROSAIL RTM approach on Landsat 8 data could be utilized to accurately estimate LAI at regional scale which could aid in rapid assessment and monitoring of the rangeland resources. / Environmental Sciences / M. Sc. (Environmental Science)

MSc (Botany) / Department of Botany / Plant populations are determined by many elements that impose demographic filters. Following disturbance, in African savanna woodland trees regenerate both sexually and vegetatively. Vachellia karroo (Hayne) Banfi and Galasso, family Fabaceae and sub family Mimosideae, commonly known as “sweet thorn” or “soetdoring” (Afrikaans), is a leguminous shrub common throughout Southern Africa. Its distribution range is from the Southwestern Cape northwards into Namibia, Angola, Botswana, Zambia and Zimbabwe. It is a pioneer species and has the ability to encroach rapidly into grassland grazing areas. Therefore, V. karroo is considered the most important woody invader of grasslands in South Africa. This study aimed at investigating the population biology and ecology of V. karroo in the clay waterlogged site of the Nylsvley Nature Reserve. Plant height, stem circumference, canopy cover, distance between individuals were measured and disturbance levels on individuals of V. karroo were estimated during sampling. The results showed that the population of V. karroo display an adequate growth curve thus implying that the population of V. karroo in the Nylsvley Nature Reserve is healthy and viable. More mature healthy individuals were found in September and December, than during June and July; this is obviously because early summer season, and in September most of the savanna plant species were starting to actively recover from the winter dry season. This is confirmation enough that the population of V. karroo is expanding in the clay waterlogged areas of Nylsvley Nature Reserve.

Thesis (MSc)--Stellenbosch University, 2004. / ENGLISH ABSTRACT: Bush encroachment in savannas leads to reduced diversity, productivity and profitability of rangelands. This holds important implications for the livestock and eco-tourism industries, as well as for subsistence ranchers in the South African semi-arid savannas, who depend on this vegetation type for economic and livelihood purposes. Soil moisture, nutrients, rue and herbivory are generally regarded as the principal factors determining vegetation structure and function within savannas. The factors and processes involved in the determination of the tree:grass ratio within savannas are, however, not clearly understood. We investigated the role of soil type and management strategy (cattle, game and communal grazing) in the determination of the presence and distribution of plant species in general, and on the presence and distribution of the encroaching tree species, Acacia mellifera, specifically. Both shortand long-term trends were investigated. The study area, the Kimberley Triangle, (Northern Cape Province, South Africa), was ideal for this kind of study because it has different management strategies practised on several soil types, and bush encroachment is a widespread phenomenon. Contrary to the belief that heavy livestock grazing is the main cause of increases in the tree:grass ratio, we found that soil type, through its effects on plant growth and on the presence and availability of soil moisture and nutrients, is more important in determining vegetation composition than management strategy. It was found that the various types of grazing management mainly influenced vegetation structure and function by affecting the competitive interactions between Ns-fixing woody species and non-Ns-fixing grasses. Of the soil factors affecting vegetation composition, soil texture was a good indicator of the physical conditions for plant growth in an area, and also of the presence and availability of soil moisture and nutrients. We found that sand and clay soils are relatively resistant to bush encroachment as compared to loamy-sand and -clay mixes. This is because woody growth is impaired in the first-mentioned habitats by soil texture, soil moisture regimes and heavy utilisation. In soils with combinations of loam, sand and clay, soil texture and moisture are not limiting to woody growth and if the repressive competitive effect of grasses on woody vegetation is removed, opportunities are created for recruitment of woody species and encroachment. Additionally, rockiness increases soil moisture infiltration. In the study area, woody species, and specifically A. mellifera, are associated with these areas. We suggest that in the study area, rocky areas are naturally encroached. This is in agreement with the patchdynamic approach to savanna vegetation dynamics. Because soil moisture is such an important factor in the determination of the tree:grass ratio in the study area, we further suggest that in above-average rainfall years, when soil moisture conditions are optimal for woody seed germination, establishment and growth, heavy grazing should be avoided, as it would provide the opportunity for encroachment. Keywords: Bush encroachment; N2-fixing species; game, cattle, communal grazing; soil texture; soil moisture; soil nutrients; Acacia mellifera; tree-grass competition. / AFRIKAANSE OPSOMMING: Bosindringing in savanna gebiede het verlaagde diversiteit, produktiwiteit en winsgewendheid van natuurlike weivelde tot gevolg. Behalwe dat dit die Suid-Afrikaanse vee- en ekotoerisme bedrywe raak, is verskeie bestaansboerderye afhanklik van die plantegroei-tipe vir oorlewing. Grondvog, grondvoedingstowwe, vuur en beweiding word algemeen aanvaar as die belangrikste faktore wat die struktuur en funksie van savannas bepaal. Daar is egter nog nie duidelikheid oor die prosesse wat betrokke is in die bepaling van die boom-gras verhouding in savannas nie. In dié studie het ons ondersoek ingestel na die invloed van grond tipe en beweidingstrategieë (beweiding deur beeste, wild, of kommunale beweiding) op die algemene verspreiding van verskillende plant spesies, en ook op die van die indringer spesie, Acacia mellifera. Beide kort- en lang-termyn patrone is ondersoek. Die studiegebied, die Kimberley Driehoek in die Noord-Kaap, Suid-Afrika, was ideaal vir so 'n ondersoek omdat verskillende beweiding strategieë op verskeie grond-tipes beoefen word maar bosindringing steeds 'n algemene verskynsel in die gebied is. Ons bevinding was dat, ten spyte van die feit dat daar oor die algemeen geglo word dat swaar beweiding die hoof oorsaak van bosindringing is, grond-tipe belangriker is in die bepaling van die spesie-samestelling van 'n gebied. Dit is as gevolg van die feit dat grond-tipe die groei van plante beïnvloed deur die teenwoordigheid en beskikbaarheid van grondvog en -nutriënte te bepaal. Die verskillende beweidingstrategieë beïnvloed meerendeels die kompetatiewe interaksies tussen N2-bindende hout-agtige spesies teenoor nie-Nj-bindende gras-agtige spesies. Grond-tekstuur was 'n goeie aanduiding van die algemene toestande vir plantegroei en ook van die teenwoordigheid en beskikbaarheid van grondvog en -nutriënte. Ons het bevind dat sand en klei grond, relatief tot leem, sand en klei kombinasies, weerstandbiedend is teen bosindringing as gevolg van die tekstuur, water-regimes en swaar beweidings vlakke wat op die grond-tipes voorkom. Omdat grond-tekstuur en grondvog nie beperkend is op die leem-, sand- en klei-grond kombinasies nie, kan bosindringing maklik hier voorkom as die onderdrukkende effek wat grasse op houtagtige plantegroei het, deur swaar beweiding verwyder word. A. mellifera is ook oor die algemeen met klipperige gebiede geassosieer omdat klipperigheid lei tot verhoogde infiltrasie van grondvog. In die studie-gebied is klipperige areas dan ook van nature ingedring deur A. mellifera. Dit stem ooreen met die siening dat savannas bestaan uit "laslappe" van verskillende plantegroei (patch-dynamic approach). Dit was duidelik dat grondvog 'n belangrike bepalende faktor is in die bepaling van die boom-gras verhouding in die studiegebied. Ons stel dus voor dat in bo-gemiddelde reënval jare, swaar beweiding vermy moet word, omdat houtagtige saailinge gedurende die tydperke maklik kan ontkiem en vestig juis omdat grondvog dan nie beperkend is nie. Sleutelwoorde: Bosindringing; N2-bindende spesies; bees, wild, kommunale beweiding; grond tekstuur; grond-vog; grond-nutriënte; Acacia mellifera; boom-gras kompetisie.