The optimal utilization of woody plants in savannas depends on a better understanding of plant-animal interactions. Morphological structures of plants (e.g. spines, thorns and prickles) are reported to act as plant defences against herbivory and have been shown to affect the foraging efficiency of mammalian herbivores. Moreover, condensed tannins (CTs) have proved to mediate forage intake and energy uptake by browsers, due to their digestibility reducing effects. In two short-term pen experiments (18 days and 6 days long), we investigated the effects of plant morphological characteristics (i.e., thorn presence/absence, spine length, amount of leaves per branch) and leaf chemistry (i.e. nitrogen and CT content) on the bite sizes of indigenous goats. We hypothesised that bite size would be higher for spineless plants and will increase along with CT intake. We expected spine length to negatively affect bite size, and the amount of leaves to have a positive effect on bite sizes. Nitrogen content was predicted to increase bite sizes while CTs were expected to have an opposite relationship. For the first experiment, we orally dosed 9 goats with (i) 20 g of condensed tannins extract dissolved in 50 mL of water (high tannin exposure), another 9 goats (ii) with 20 g of polyethylene glycol dissolved in 50 mL of water, and the last group of 9 goats (iii) with 50 mL of water (control) before offering forage species. Goats were offered branches of Vachallia nilotica, V. tortilis, and V. Karroo and their foraging behaviour was recorded. These branches were cut into pieces of approximately 50 cm that all had more or less the same size and structure, and the exact weight was determined using a scale with a precision of 0.01 g. All goats were fed ad libitum Lucerne hay (Medicago sativa). The second experiment using the same goats involved removal of thorns from branches and comparing bite sizes between branches within species. Goats were conditioned for a week prior to the feeding trial to re-adapt to the diets. Feeding time, number of bites taken, and weight loss of the branches was measured, in order to calculate bite size and bite rate. Bite size [g bite-1] of the goats was calculated. As predicted, CT contents of the plants and proportion of leaves per branch were found to positively affect bite size (Pearson: R > 0.6, N = 453, p < 0.001), and nitrogen content of the leaves and length of the spines were negatively correlated to bite size (Pearson: R < -0.2, N = 453, p < 0.001). The main conclusions were that (1) spine-absence is effective in reducing leaf loss to goats and, (2) supplemental treatment with CT did not affect bite size, which indicated that bite size was not affected by total CT intake in the short term. This study extends our understanding of browse-browser interactions, which represent important input to goat production and range forage species management.

Currently there are only a limited number of introduced forage species suitable for dryland farming under water-limited agro-ecosystems. The projected climate change scenarios for South Africa (SA) however, indicate a general trend of becoming hotter and drier. Recently, a number of indigenous Indigofera and Lessertia species were prioritised for further evaluation as forage crops within these water-limited agro-ecosystems. It is important however, to determine how future bioclimatic conditions could influence the distributions of these species. This in turn, will influence prioritisation of germplasm collections which, in turn, could influence the agronomic potential of these species. The DOMAIN model in DivaGIS was used to determine the climate adaptation, as well as the most limiting bioclimatic factors, limiting the distributions of the selected legume species, using the 19 bioclimatic variables of the worldclim climate database version 1.4 at the resolution of 2.5 ARC minutes. The adaptation zones were overlaid onto the different general soil classes of SA, and clipped to contain only soils on which the legumes occur. The remaining areas were then calculated as the adaptation zones and changes in distributions of each legume species between current and future bioclimatic scenarios were determined. I. heterotricha and I. meyeriana as well as all Lessertia species showed an overall increase, while I. alternans and I. nigromontana showed an overall decrease in their total predicted adaptation ranges throughout SA. All species however, displayed decreases or no changes in their highly adapted ranges from current to future bioclimatic conditions. Under the projected future bioclimatic conditions of SA, the distributions of I. meyeriana, L. diffusa and L. excisa were found to be limited primarily by precipitation variables. The distributions of all other Indigofera species as well as L. pauciflora and L. frutescence were found to be limited primarily by temperature variables, while the distributions of L. depressa and L. inflata are equally influenced by temperature and precipitation variables. Due to the variation in bioclimatic variables limiting the potential distributions and suitability of the different legume species, germplasm collections should include collections over a wide range of bioclimatic conditions in order to obtain a large variation of accessions, adapted to different agro-ecological conditions. These genetic resources will be the raw materials needed for breeding accessions and cultivars of Indigofera and Lessertia species suitable for specific agro-ecological conditions.

Veld and crop residues are main source of ruminant livestock feed in rural Africa. Veld grasses have low quality during winter season and do not meet the nutrient requirements for livestock year-round. Commercial feed is too costly and unpopular in resource poor livestock production system. Therefore the use of browse plants to supplement protein during the dry season to grazing livestock in small-scale farming systems is worthy of consideration. The study evaluated the crude protein (CP) content of browse plants as protein source for ruminants in the Eastern Cape. Nineteen browse plants (Vachellia karoo, Calpurnia aurea, Oleaeuropaea subsp. Africana, Schotia latifolia, Grewia occidentalis, Rhus pyroides, Cordia rudis, Salanum mauritianum, Diospyros dichrophylla, Vepris lanceolata, Prunus persica, Ziziphus mucronata, Zanthaxylum capense, Mimusops obovata, Cussonia spicata and Diosporus lyciodis) were collected from Goso Administrative area under the Ngquza Hill Local Municipality in June 2014 and brought to Dohne Agricultural Development Institute. Leaf fractions were oven dried at 400C for 48 hours and later grounded in a Willey mill to pass through a 1 mm sieve prior to chemical analysis to determine their respective protein contents. The CP was analysed using the procedures as outlined by Association of Analytical Chemistry (AOAC, 2002). The CP content on dry matter basis of leaf fraction ranged from 6 to 48% . There were 21% of the leaf fraction that had a CP content of < 7% (Vachellia karoo, Grewia occidentalis, Mimusops obovata, Cordia rudis), 73.7% of browse plants had CP within the range of 7.25 to 20%, and 5.3% had a CP content of above 20%. These findings indicate that the leaf fraction is the most suitable plant part use as protein source for the low nutritive value of veld in winter. It is noteworthy that the leaf of Salanum mauritianum had the highest CP content above 20%. Further research is necessary to determine the tannin content, acceptability and digestibility of the browse plants and its protein. Effect of browse plants leaves on animal performance as protein supplement warrants further study.

Natural veld has long been acknowledged to play an important role in the South African beef industry. According to the Department of Agriculture, approximately 80% of the South African agricultural land is only suitable for extensive grazing. Therefore, grazing of beef cattle in these areas, are a practical method of meat production. The aim of the study was to evaluate the effect of high utilization grazing (HUG) and controlled selective grazing (CSG) systems over a four-year period (2011/2012 – 2014/2015) on veld condition in the Sourish Mixed Bushveld. The experimental herd consisted of 92 Bonsmara cows, divided in four sub-herds consisting of 23 cows each. With the CSG treatment, animals were moved from one camp to the next when approximately 30% of available grass dry matter was utilized whereas with the HUG treatment, animals were moved when approximately 60% of available grass dry matter was utilized. Veld evaluation was done during each consecutive growing season (October/November). Both the area-based method and point-based method were used to determine: Basal cover percentage; Total canopy cover percentage; Proportional canopy cover percentage of grasses; Standing biomass (kg/ha); Decreaser grass species contribution (%); and Veld condition score (number out of 1000). The data was subjected to an appropriate factorial analysis of variance (ANOVA) using the repeated measurements as a Sub-Plot Factor. The Shapiro-Wilk’s test was performed on the standardized residuals to test for deviations from normality. In cases where significant deviation from normality was due to skewness, outliers were removed, until normal or symmetrically distributed. The student's t- Least significant difference (LSD) was calculated at a 5% significance level to compare means of significant source effects. All the above data analyses were performed using SAS version 9.3 statistical software. The basal cover percentage of the HUG treatment decreased significantly from 2011/2012 to 2014/2015 and the basal cover for the CSG treatment decreased slightly, although not significantly. The reason for the significant decrease in grass plant basal size for the high utilization grazing treatment may be due to the higher defoliation of grass leaves. The average standing biomass for the HUG treatment varied from 2394 kg/ha in 2012/2013 to 2994 kg/ha in 2014/2015 and for the CSG treatment from 2514 kg/ha in 2012/2013 to 3044 kg/ha in 2014/2015. Over the four-year period, a significant difference was found between HUG and CSG (P=0.001) mean veld condition scores, but the differences over years was not significant (P = 0.51). The decreaser grass species increased slightly from the first evaluation in 2011/2012 to the last evaluation in 2014/2015, for both treatments. These results indicate that neither the HUG nor CSG had a negative impact on the veld in relation to the decreaser grass species composition. It is also an indication that the veld was not under grazed or overgrazed. Note: This research is financially supported by Red Meat Research and Development South Africa (RMRD SA) and THRIP

Old lands in the Eastern Cape Province suffer from low pasture productivity, brought about by a lack of diversity in the forage species available. Previous trial work identified a suite of annual and perennial legume pastures to fill the winter feed gap and significantly increase in animal production. However, to ensure long term sustainability these pastures, need a good seeding rest every three years. For the temperate species, this is typically during the summer season when maize is grown.  The aim is to develop an intercropping system that could utilize any excess nitrogen produced by the pasture legumes for maize production, while affording the legumes an opportunity to set seed. At two communities (Sinqumeni and Kubedlana) each, a plot (15 m X 80 m) with an even stand of legumes was selected in an existing grass and legume pasture. Herbage production was determined by cutting 1 m² quadrats during November and May of each year. A 250 g grab sample was collected per plot for quality analysis purposes. To determine soil nitrate-N levels, five soil sample cores were collected at random at 4 weekly intervals commencing in November 2016. The samples were pooled per plot for analysis purposes. Both the herbage and the soil data indicate that there was no excess Nitrogen available for the maize crop. From the soil data, a calculated 9 kg N/ha was available while the calculation of plant available Nitrogen based on herbage data show that there was no Nitrogen available for the planted maize crop.There are a number of factors that can contribute to this, but in our opinion the utilization of the pasture during the winter is the biggest factor. High quality forage is grazed by animals and the Nitrogen is then deposited in urine and dung in other areas such as kraals and around water points.

Soil nutrients play a vital role in tree-grass interactions within an ecosystem. Soil pH, macro and micro elements are determinants of vegetation productivity. Encroachment of woody plants in an ecosystem tends to outcompete grasses. Competition between of grasses and woody plants impact the soil and a deficiency of soil minerals, can lead to deficiency in the forages and subsequently of the livestock. The effect of bush encroachment on soil nutrient content is well documented, but there is limited data available that clarifies the impact of bush encroachment on soil properties such as soil pH, and macro and micro elements in communal grazing lands. The objective of the study was to assess the effects of bush encroachment on soil nutrient content at Sheshegu communal rangeland. The data was collected from four different bush densities: Scattered (15%), Moderate (30%), Mixed (45%) and Dense (60%), with these densities replicated three times. A plot of 100 m x 50 m was demarcated in each replicate. Two soil samples per bush density were sampled randomly using soil auger to a depth of 30cm. A total of 12 soil samples were collected, oven dried for 48 hours at 60°c, and analysed for N, P, K, organic Carbon (OC), Cu, Fe, Mn, Zn and soil pH. Inductively Coupled Plasma (ICP) was used to determine the concentration of P, Mg, P, Zn, Mn, Fe and Cu using 1% of critic acid and 0.02M DI-ammonium Ethylene-diamine-tetra-acetic acid. The standard Kjeldahl method was used to determine N concentration. The Walkley-black technique was used to determine OC. Soil pH was determined in 1M potassium chloride 2.5 KCL: 1 soil. The dense treatment (60%) indicated high nitrogen content compared to the scattered (15%), moderate (30%) and mixed (45%) treatments, respectively. There was a low concentration of OC at the mixed (45%) treatment, but the concentration of OC revealed no significant difference (p > 0.05) between the scattered (15%), moderate (30%) and dense (60%) treatments. These findings showed that the concentration of P and Fe in the soil decreases with an increase in bush density. Soil pH indicated a positive relationship with increase in bush density. This implies that an increase in bush encroachment could have an impact on soil nutrient content, particularly, P, soil pH and OC. Proper veld management practices are recommended to mitigate bush encroachment, in order to improve rangeland condition in Sheshegu communal rangeland.

 

Accurate and near-real time estimation of herbaceous aboveground biomass (AGB) at farm level is crucial for monitoring utilisation of pasture and proactive management of stock in semi-arid rangelands. Despite its importance, remote sensing has been rarely used by range ecologists and managers in Zimbabwe. This study aimed to assess the performance of classical multispectral vegetation indices (MVIs) when either singly regressed with measured herbaceous AGB or combined with other visible spectral bands in predicting herbaceous AGB in a Colophospermum mopane tree-shrub savanna. Field herbaceous AGB and corresponding Landsat 8 Operational Land Imager (OLI) visible spectral data were collected during the 2016-17 rainy season. Relationships between measured AGB and classical MVIs and extended models of MVIs combined with other visible spectral bands were analysed using bootstrapped simple and stepwise multiple linear regression functions. When MVIs were singly regressed with measured AGB, ratio-based indices yielded the highest r2 value of 0.64 followed by soil-adjusted indices (r2 = 0.61) whilst atmospheric-corrected MVIs showed the lowest r2 of 0.58 (p = 0.00). A significant improvement in herbaceous AGB estimation was obtained by using a combination of classical MVIs and other visible spectral bands. Soil-adjusted MVIs showed the greatest increase (44-46 %) in r2 whilst atmospheric-corrected and ratio-based MVIs poorly improved (less than 5 %). The findings demonstrate that combining MVIs with Landsat 8 optical spectral bands, especially green band provides the best models for estimating AGB in C. mopane savanna rangelands. These findings emphasise the importance of testing band-MVI combinations when developing models for estimating herbaceous AGB.

Bush encroachment is a form of land degradation that is noticeable globally, and predominantly in semi-arid areas. Bush encroachment in savannas can alter carbon (C) and nitrogen (N) levels over the long-term, which could have environmental implications. The impacts of bush encroachment on soil organic carbon (SOC), soil total nitrogen and soil aggregate stability were assessed at simulated different levels (12% 25%, 50%, 75% and 100%) of encroachment at Honeydale farm in Alice, Eastern Cape Province. Soil samples were collected at three depths (20-40 cm, 41-60 cm and 61-80 cm) in all treatments and were separated into two for chemical analysis and for aggregate stability, respectively. Samples for aggregate stability were air dried and passed through a 5-mm sieve. Visible organic materials and debris were discarded. Aggregate stability was measured according to Le Bissonnais (1996). Thereafter, three replicate soil samples were immersed in 50 mL deionized water and the material was gently transferred to a 50 μm sieve that immersed in ethanol to separate it into 50 μm fragments. The remaining >50 μm fraction was oven dried and sifted on a stack of sieves of 2000, 1000, 500, 250 and 50 μm pore size. The weight of each fraction was then measured, and soil fraction < 50 μm was calculated and expressed as the mean weight diameter (MWD). The results indicated no consistency and less signs of soil aggregate instability, however treatments with less bush density had higher figures of mean weight diameter. Soils from all treatments had an acidic reaction at all sampled depths range. The exchangeable Ca2+, which is said to promote aggregation, dominated the exchange complexes of the soils. Soil organic carbon (SOC) content was less than 2% in all samples and decreased with depth. The average SOC content for the soil at the site ranged from 0.32 to 1.17 % and it is suggested that when SOC is at 2% is considered to be in a critical level, below which soil structural stability will suffer a significant decline. Bush density had slight influence on soil aggregate stability.

Due to their popularity for tourism, giraffes are being introduced outside their historical distributions into Thicket areas within the Little Karoo region of the Western Cape. However, information regarding the activity budgets and diet selection of these giraffes is lacking. Therefore, this study aimed to achieve three objectives; to determine the diurnal activity budget, diet composition and browsing levels, and the estimated browsing capacities of extralimital giraffe. The study was conducted on two privately owned farms, namely Kareesbos Private Game Reserve and Tsumkwe Private Game Reserves. Observations were completed on both study sites using the interval scan method, whereby observations were conducted on all visible individuals from 6 am – 6 pm for four days every 3 months (winter, spring, summer and autumn). Observations found browsing to be the most dominant activity displayed by both populations, with walking and rumination being the second and third most dominant activity. Females browsed more than males in both study sites. Dietary selection showed to be contrary to that found within their natural ranges where Vachellia and Senegalia spp. are prominent. Four species (Pappea capensis, Portulacaria afra, Euclea undulata and Searsia longispina) and five species (Searsia longispina, Euclea undulata, Pappea capensis, Vachellia karroo and Grewia robusta) comprised approximately 90% and 80% of the giraffes’ diet in Kareesbos and Tsumkwe respectively. The importance of flower bearing species (Lycium spp. and Rhigozhum obovatum) increased during the spring and summer seasons in both study sites. In addition to the diet selection, browsing by both giraffe populations was shown to occur mostly below 2 m. Estimated browsing capacities for Kareesbos and Tsumkwe were 25 and 21, and 107 and 88 ha per giraffe, respectively, for the respective browse height strata of less than two metres and five metres. It is suggested that browsing capacities of less than two metres be considered when stocking giraffe and the number of individuals adjusted accordingly on each farm, due to the continuous low browsing of giraffe at less than two metres. Furthermore, results indicate that these giraffes have adapted to take advantage of forage available in ecosystems outside their natural ranges. Low foraging heights suggests possible niche overlap with other browsers, which may result in increased competition for food when it becomes limited. Long term ecological monitoring of extralimital populations and appropriate management procedures are therefore required to avoid the displacement and degradation of indigenous fauna and flora within the Little Karoo.

Old lands in the Eastern Cape Province suffer from low pasture productivity, due to low diversity in forage species. The ECCAL project identified a suite of pasture legumes to fill the winter feed gap and significantly increase animal production. However, to ensure long term sustainability these legumes, need a seeding rest every three years. For the temperate species, this is typically during the summer season when maize is grown. This created an opportunity to develop a system that could utilize any excess nitrogen produced by the pasture legumes for maize production, while affording the legumes an opportunity to set seed. The cover on these rehabilitated old lands consist of a mixture of grass and legumes which could have a negative influence on the germination and establishment of maize seedlings. A herbicide trial was therefore designed with the aim of supressing grass and legume growth but allowing the legumes to survive. A plot (15 m X 80 m) with an even stand of legumes was selected in a rehabilitated old land. In November 2016 eight different herbicide treatments were applied in a randomized block design with four replications. Species composition and stand density was determined by conducting a frequency count annually during November and May. During the last week of October 2017 the entire plot was burnt down by a runaway veld fire. This afforded us the opportunity to assess the effect that a late winter or spring burn will have on the survival rate of the different legumes in a rehabilitated old land. Data collected in November 2017 (3 weeks after the fire) indicated a reduction in legume numbers ranging from 42% in Arrowleaf clover (Trifolium vesiculosum) to 23% in Lespedeza (Lespedeza cuneata) plants, compared ton the May 2017 count.  It can be concluded that legumes have the ability to survive a late winter or spring fire regardless of the application of herbicides during the previous spring.
 
 
 
 

Goats are raised in conditions which do not allow them to achieve their maximum performance, nor express their genetic potential. They are raised on poor quality feeds, with low energy and protein content. For goats to perform to their full potential as small ruminants, it requires highly nutritious feed, especially during high producing periods. However, supplementation can be done using fresh green feeds, protein blocks or vitamin supplements as well as traditional food crops residues. This study outlines the knowledge and perceptions of goat farmers concerning feeding sweet potato vines to goats. This study was conducted at KwaMthethwa community area, KwaMbonambi local Municipality, KwaZulu-Natal, which consist of seven wards. Surveys were administered through structured questionnaires and 15 households were interviewed per ward, making a sample size of 105 participants. Participants were randomly selected at their willingness to participate in an interview. Data was analysed using frequency procedure of SPSS 2015. Results of this study showed that 56.2% of the respondents were males and 43.8% were female, demonstrating that male dominion in the farming sector, still exists. The results also revealed that 71% of households keep goats for socio-cultural purposes such as lobola, traditional functions and status, whereas 15% and 9% is kept for income generating purposes and consumption, respectively. The majority of farmers (71.4%) were aware of, and do, practice some sort of supplementary feeding. The source of feed being bought was 34%, of which 26.7% was sourced from food crop residues and 9% from the branches of indigenous trees. Farmers (72.4%) cultivate sweet potatoes for different uses such as generating income (53.3%) or for consumption (27.5%). A further 8% use it for both consumption and income, whilst 16% do not cultivate sweet potatoes. Sweet potato vines were being discarded as waste or burnt (32.4%), left on the field as green manure (25.7%), conserved as propagation material (8.6%), and fed to livestock by 7.6% respondents, respectively. However most of the participants (78.1%) rejected the use of sweet potato vines as feed to livestock. Their assumption behind rejecting sweet potato vines needs to be studied further, since they believe that the vines cause fatal diarrhoea to goats. Domination of males in livestock production sector is still high and the findings concur with previous studies that indicate that small holding farmers keep livestock for socio-economic purposes. Most farmers seem to know about supplementation, however they have little understanding concerning the concept of supplementation, since most were giving maize to goats every afternoon for goats to come back for kraaling in the afternoons without herder. Government still has a major role to play to ensure more women partake in agricultural sector. Furthermore, there is a need to develop farmers in goat farming through training and workshops concerning supplementary feeding, for farmers to be able to supplement in a profitable manner.

Grassland biomes globally are facing increasing anthropogenic pressure as human populations increase, resulting in an increased need for the resources that grasslands provide. Likewise, South African grasslands are facing increased habitat loss and fragmentation, and have thus become severely threatened, making them a priority for conservation efforts. There are growing concerns about the potential ecological consequences of the increased impact of human activities, especially loss of biodiversity. The knowledge and understanding of how resource availability and environmental distress affect plant diversity is key to proper conservation and management of grassland diversity. Our study aimed at providing more understanding on the interactions of environmental resources and biodiversity, by comparing plant diversity between wildlife-protected areas and adjacent non-protected grazing lands in the South African Grassland Biome. Three nature reserves were chosen around the Gauteng Province, namely Roodeplaat Nature Reserve (RNR) in Tshwane, Suikerbosrand Nature Reserve (SNR) in Heidelberg, and Abe Bailey Nature Reserve (ANR) in Carletonville. We selected three sites within RNR, SNR and ANR, and paired these with adjacent sites in private farming/grazing areas adjacent to the nature reserves. At each site, species composition and plant diversity were determined with the use of 50m x 20m Modified-Whittaker plots, making a total of eighteen plots at the nine paired sites. Animal grazing was visually estimated at each site, based on the amount of plant defoliation and trampling, soil disturbance, amount of dung, and animal footpaths. Three grazing intensities were distinguished: low/no grazing, moderate grazing, and heavy grazing. Our results for the 1000 m2 plots indicate similar species richness (S) at two paired sites at ANR, but the third site had higher S in the adjacent private grazing area than in the nature reserve. This moderately grazed site also had the highest component of grass species, both inside and outside the nature reserve. At SNR, two moderately grazed sites had higher S in the adjacent private grazing area. There was no difference in S on the third site, but the adjacent private grazing area had a higher proportion of grass species. Two paired sites at RNR had similar S within the reserve and in the adjacent private grazing area, but one of the sites had a higher proportion of grass species. The third paired site had high S in the nature reserve compared to the adjacent private grazing area, however the site in the adjacent private grazing area had a much higher proportion of grass species. With regard to grazing, the moderately grazed and overgrazed sites had in general a higher number of palatable grass species as compared to the low grazing areas. The low grazing and heavy grazing sites had a higher cover abundance of the unpalatable and this was evident with the increase in species such as Hyparrhenia hirta and Cymbopogon caesius. The results suggest that species richness in both wildlife-protected and non-protected private grazing area sites is influenced by grazing intensity and this leads to lower S in over-utilised and underutilized grasslands, compared to higher richness at moderately utilised sites.