Effect of dietary Mucuna pruriens seed meal on the serum lipid profile and bone quality traits of broiler chickens
- Authors: Hempe, Chumani
- Date: 2019
- Subjects: Legumes Lipoproteins
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/17213 , vital:40866
- Description: This study aimed at evaluating the effects of heated Mucuna pruriens seed meal on the blood serum lipid profile and bone quality parameters of broiler chickens. A total of 120 day-old unsexed Cobb 500 broiler chicks were randomly allocated for the 4 treatment diets (T1, T2, T3 and T4) containing 0, 10, 15, and 20% heated Mucuna pruriens seed meal. The birds were reared on wood shaving floor pens for 35 days. Birds were humanely slaughtered by severing the jugular vein using a sharp knife. Blood was collected into a tube, centrifuge and analyzed for a lipid profile analysis (total cholesterol (TC), high density lipoproteins (HDL), low density lipoproteins (LDL), triglycerides (TCD) as well as aspartate transaminase (AST) and alanine transaminase (ALT)) by enzymatic diagnostic kits (Diasys diagnostic kits). Both drumsticks severed using a sharp knife. After boiling in the water at 100°C for 10 mins the tibia bones were taken, air-dried for 96 hrs at room temperature and measured for weights. Bone length, strength, moisture and ash content as well as morphology were measured. The blood lipid profile of birds fed T1 was high (P < 0.05) for all the parameters measured, while T3 had the lowest (P < 0.05) profile. T2 and T4 had the same profile. ALT was found to be highest (P < 0.05) for T4 (5.27 ± 0.207/L) and lowest (P < 0.05) for T2 (4.67 ± 0.333/L) while AST was highest (P < 0.05) for T2 (320.13 ± 28.851/L) and lowest (P < 0.05) for T3 (208.47 ± 10.059/L). Treatment diet 3 (T3) was observed decreased bone quality among all diets; with the lowest (P < 0.05) bone strength (188.55 ± 12.039N), bone length (8.59 ± 0.131cm) and morphology (1.69 ± 0.058). Mucuna pruriens seed meal inclusion in broiler diets reduces blood serum lipids and 20% inclusion levels have no adverse effects on the bone quality attributes
- Full Text:
- Date Issued: 2019
- Authors: Hempe, Chumani
- Date: 2019
- Subjects: Legumes Lipoproteins
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/17213 , vital:40866
- Description: This study aimed at evaluating the effects of heated Mucuna pruriens seed meal on the blood serum lipid profile and bone quality parameters of broiler chickens. A total of 120 day-old unsexed Cobb 500 broiler chicks were randomly allocated for the 4 treatment diets (T1, T2, T3 and T4) containing 0, 10, 15, and 20% heated Mucuna pruriens seed meal. The birds were reared on wood shaving floor pens for 35 days. Birds were humanely slaughtered by severing the jugular vein using a sharp knife. Blood was collected into a tube, centrifuge and analyzed for a lipid profile analysis (total cholesterol (TC), high density lipoproteins (HDL), low density lipoproteins (LDL), triglycerides (TCD) as well as aspartate transaminase (AST) and alanine transaminase (ALT)) by enzymatic diagnostic kits (Diasys diagnostic kits). Both drumsticks severed using a sharp knife. After boiling in the water at 100°C for 10 mins the tibia bones were taken, air-dried for 96 hrs at room temperature and measured for weights. Bone length, strength, moisture and ash content as well as morphology were measured. The blood lipid profile of birds fed T1 was high (P < 0.05) for all the parameters measured, while T3 had the lowest (P < 0.05) profile. T2 and T4 had the same profile. ALT was found to be highest (P < 0.05) for T4 (5.27 ± 0.207/L) and lowest (P < 0.05) for T2 (4.67 ± 0.333/L) while AST was highest (P < 0.05) for T2 (320.13 ± 28.851/L) and lowest (P < 0.05) for T3 (208.47 ± 10.059/L). Treatment diet 3 (T3) was observed decreased bone quality among all diets; with the lowest (P < 0.05) bone strength (188.55 ± 12.039N), bone length (8.59 ± 0.131cm) and morphology (1.69 ± 0.058). Mucuna pruriens seed meal inclusion in broiler diets reduces blood serum lipids and 20% inclusion levels have no adverse effects on the bone quality attributes
- Full Text:
- Date Issued: 2019
Effect of grazing systems on growth performance, carcass characteristic, meat quality of beef cattle and economic performance of the enterprise
- Authors: Dayimani, Siyavuya
- Date: 2019
- Subjects: Grazing -- Environmental aspects Meat -- Quality
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/17294 , vital:40873
- Description: Under conditions of adaptive Holistic Management™, we tested whether rotational grazing approaches such as holistic planned grazing (HPG) and four-camp grazing (FCG) influence growth, carcass and meat quality attributes of beef steers compared to continuous, seasonlong (SLG) grazing. Thirty steers (250 ±19.56 kg, 16-17-month-old) were randomly assigned to SLG, HPG and FCG systems (10 steers/ treatment) at a private farm in Cedarville, Eastern Cape Province, South Africa where grazing was managed using an adaptive approach involving inter alia planned grazing charts, seasonal movement of cattle, and supplemental feeding. The steers were slaughtered at 40-41 months for carcass measurements and longissimus et lumborum was sampled for meat quality analyses. All the grazing systems had comparable live weights and average daily gain. All the investigated carcass attributes were also similar across grazing treatments (P > 0.05). Meat from steers on the SLG system had higher (P<0.05) drip loss, protein and fat contents and lower (P <0.05) shear force, moisture and ash contents than the two rotational grazing approaches (HPG and FCG). The HPG had higher (P < 0.05) total running and capital costs compared to SLG and FCG grazing systems. The SLG system had greater (P ≤ 0.05) gross and net income compared to FCG and HPG systems, respectively. Overall, all the grazing systems had comparable growth performance, carcass characteristics and meat quality but rotational grazing systems had lower gross income relative to the SLG grazing system.
- Full Text:
- Date Issued: 2019
- Authors: Dayimani, Siyavuya
- Date: 2019
- Subjects: Grazing -- Environmental aspects Meat -- Quality
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/17294 , vital:40873
- Description: Under conditions of adaptive Holistic Management™, we tested whether rotational grazing approaches such as holistic planned grazing (HPG) and four-camp grazing (FCG) influence growth, carcass and meat quality attributes of beef steers compared to continuous, seasonlong (SLG) grazing. Thirty steers (250 ±19.56 kg, 16-17-month-old) were randomly assigned to SLG, HPG and FCG systems (10 steers/ treatment) at a private farm in Cedarville, Eastern Cape Province, South Africa where grazing was managed using an adaptive approach involving inter alia planned grazing charts, seasonal movement of cattle, and supplemental feeding. The steers were slaughtered at 40-41 months for carcass measurements and longissimus et lumborum was sampled for meat quality analyses. All the grazing systems had comparable live weights and average daily gain. All the investigated carcass attributes were also similar across grazing treatments (P > 0.05). Meat from steers on the SLG system had higher (P<0.05) drip loss, protein and fat contents and lower (P <0.05) shear force, moisture and ash contents than the two rotational grazing approaches (HPG and FCG). The HPG had higher (P < 0.05) total running and capital costs compared to SLG and FCG grazing systems. The SLG system had greater (P ≤ 0.05) gross and net income compared to FCG and HPG systems, respectively. Overall, all the grazing systems had comparable growth performance, carcass characteristics and meat quality but rotational grazing systems had lower gross income relative to the SLG grazing system.
- Full Text:
- Date Issued: 2019
Effect of single-walled carbon nanotubes on various properties of reclaimed rubber/natural rubber blends
- Authors: Gumede, Jabulani Innocent
- Date: 2019
- Subjects: Carbon nanotubes , Nanotubes Nanostructured materials Polymers
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/39675 , vital:35347
- Description: Mechanical and thermal processes are the preferred methods used to reclaim rubber from waste tyres. However, these methods produce reclaimed rubber (RR) with poor quality and lowered properties compared to virgin rubber. As a result, RR is commonly blended with virgin natural rubber (NR) in order to enhance its quality and properties to widen its application scope. The use of RR in the production of new rubber products is however limited to only about 1-10%. This is because the amounts of RR above this percentage negatively affect the properties of the resultant RR/NR blends. Researchers have used various conventional and nonconventional fillers such as carbon black and maize stalk fibres, respectively, in an attempt to enhance the properties of RR/NR blends. One main disadvantage with the use of these fillers is the need for high quantities to achieve desired properties. Several researchers have therefore focused on developing other reinforcing fillers. The fillers that have recently attracted a lot of attention of the present researchers include single-walled carbon nanotubes (SWCNTs), which are very small tubes solely made up of carbon molecules. This is because SWCNTs offer many impressive properties, and even small quantity of these nanofillers is sufficient to achieve desired properties due to their small size, large surface area and high aspect ratio. The main challenge with the use of SWCNTs however is getting them to disperse well in the rubber matrix. Therefore, their surface is either modified and then mixed with rubber or a dispersant is used to disperse them in the rubber matrix, forming a masterbatch. The common nanomaterial masterbatch product for applications in rubber is the TUBALLTM RUBBER (TR), which is a masterbatch product consisting of 1 wt. % TUBALL™ SWCNTs dispersed, using 1.5 wt.% sulfonol as a dispersant, in 97.5 wt.% NR (from latex Taytex FA). This product is a versatile masterbatch designed to enhance the mechanical, chemical, electrical and thermal properties of rubbers since it contains SWCNTs. However, the performance of this product has not been explored in the cases where recycled/reclaimed rubber is used. The main aim of this project was therefore to explore the effect of SWCNTs (via using TR) in the formulations of RR/virgin NR blends. This project was divided into two parts; (i) optimisation of SWCNTs based on tensile properties of specific formulations of compounds. These compounds were prepared by mixing various amounts of TR and NR using a Banbury internal mixer and two-roll mill. (ii) the effect of the addition of SWCNTs from the optimum SWCNTs/NR masterbatch compound on curing, mechanical, aging, dynamic mechanical and thermal properties of RR/NR blends was studied. Curing properties explored include minimum torque (ML), maximum torque (MH), scorch time (ts0.5), curing time (t90) and cure rate. Mechanical properties studied are tensile strength, modulus at 100%, 200%, 300% and 500% elongation, elongation at break, hardness and resilience. Dynamic mechanical properties include stiffness, storage modulus (E'), loss modulus (E''), loss tangent delta (tanδ) and glass transition temperature (Tg). Thermal degradation (stability) was studied by means of thermogravimetric analysis (TGA) for analysing the effect of SWCNTs inclusion on RR/NR blends. The optimum SWCNTs/NR masterbatch compound contained 0.1 wt. % of SWCNTs. The results obtained for various properties indicate that SWCNTs from the optimum SWCNTs/NR masterbatch compound enhanced the curing, mechanical, aging and dynamic mechanical properties of RR/NR blends with lower RR loading. Thermal properties of RR/NR blends were found to be unaffected by the addition of SWCNTs. However, owing to their impressive properties, the use of SWCNTs as nanofillers in RR/NR blends have the greatest potential for extending the applications of RR/NR blends.
- Full Text:
- Date Issued: 2019
- Authors: Gumede, Jabulani Innocent
- Date: 2019
- Subjects: Carbon nanotubes , Nanotubes Nanostructured materials Polymers
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/39675 , vital:35347
- Description: Mechanical and thermal processes are the preferred methods used to reclaim rubber from waste tyres. However, these methods produce reclaimed rubber (RR) with poor quality and lowered properties compared to virgin rubber. As a result, RR is commonly blended with virgin natural rubber (NR) in order to enhance its quality and properties to widen its application scope. The use of RR in the production of new rubber products is however limited to only about 1-10%. This is because the amounts of RR above this percentage negatively affect the properties of the resultant RR/NR blends. Researchers have used various conventional and nonconventional fillers such as carbon black and maize stalk fibres, respectively, in an attempt to enhance the properties of RR/NR blends. One main disadvantage with the use of these fillers is the need for high quantities to achieve desired properties. Several researchers have therefore focused on developing other reinforcing fillers. The fillers that have recently attracted a lot of attention of the present researchers include single-walled carbon nanotubes (SWCNTs), which are very small tubes solely made up of carbon molecules. This is because SWCNTs offer many impressive properties, and even small quantity of these nanofillers is sufficient to achieve desired properties due to their small size, large surface area and high aspect ratio. The main challenge with the use of SWCNTs however is getting them to disperse well in the rubber matrix. Therefore, their surface is either modified and then mixed with rubber or a dispersant is used to disperse them in the rubber matrix, forming a masterbatch. The common nanomaterial masterbatch product for applications in rubber is the TUBALLTM RUBBER (TR), which is a masterbatch product consisting of 1 wt. % TUBALL™ SWCNTs dispersed, using 1.5 wt.% sulfonol as a dispersant, in 97.5 wt.% NR (from latex Taytex FA). This product is a versatile masterbatch designed to enhance the mechanical, chemical, electrical and thermal properties of rubbers since it contains SWCNTs. However, the performance of this product has not been explored in the cases where recycled/reclaimed rubber is used. The main aim of this project was therefore to explore the effect of SWCNTs (via using TR) in the formulations of RR/virgin NR blends. This project was divided into two parts; (i) optimisation of SWCNTs based on tensile properties of specific formulations of compounds. These compounds were prepared by mixing various amounts of TR and NR using a Banbury internal mixer and two-roll mill. (ii) the effect of the addition of SWCNTs from the optimum SWCNTs/NR masterbatch compound on curing, mechanical, aging, dynamic mechanical and thermal properties of RR/NR blends was studied. Curing properties explored include minimum torque (ML), maximum torque (MH), scorch time (ts0.5), curing time (t90) and cure rate. Mechanical properties studied are tensile strength, modulus at 100%, 200%, 300% and 500% elongation, elongation at break, hardness and resilience. Dynamic mechanical properties include stiffness, storage modulus (E'), loss modulus (E''), loss tangent delta (tanδ) and glass transition temperature (Tg). Thermal degradation (stability) was studied by means of thermogravimetric analysis (TGA) for analysing the effect of SWCNTs inclusion on RR/NR blends. The optimum SWCNTs/NR masterbatch compound contained 0.1 wt. % of SWCNTs. The results obtained for various properties indicate that SWCNTs from the optimum SWCNTs/NR masterbatch compound enhanced the curing, mechanical, aging and dynamic mechanical properties of RR/NR blends with lower RR loading. Thermal properties of RR/NR blends were found to be unaffected by the addition of SWCNTs. However, owing to their impressive properties, the use of SWCNTs as nanofillers in RR/NR blends have the greatest potential for extending the applications of RR/NR blends.
- Full Text:
- Date Issued: 2019
Effect of substituents on the photophysical properties and nonlinear optical properties of asymmetrical zinc(II) phthalocyanine when conjugated to semiconductor quantum dots
- Authors: Mgidlana, Sithi
- Date: 2019
- Subjects: Nonlinear optics , Quantum dots , Phthalocyanines , Zinc
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/97152 , vital:31404
- Description: Various characterization techniques have been used to characterize the synthesized asymmetrical zinc phthalocyanines (ZnPc) derivatives. Techniques include Ultraviolet-visible (UV-vis) spectrophotometry, matrix assisted laser desorption time of flight mass spectrometry (MALD-TOF MS), proton nuclear magnetic resonance (1H-NMR), elemental analysis and Fourier-transform infra-red spectroscopy (FT-IR). The complexes are covalently linked to core/shell and core/shell/shell semiconductor quantum dots (SQDs) via amide bond formation. Photophysical properties of complexes improved in the presence of semiconductor quantum dots (SQDs). SQDs contain cadmium/telluride (CdTe) as core, coated in the first shell with zinc selenide (ZnSe) or zinc sulfide (ZnS) and with zinc oxide (ZnO) in second shell. The photophysical properties of the phthalocyanine (Pc) complexes and their conjugates with SQDs are investigated in solution. Triplet quantum yields of complexes improved in the presence of semiconductor quantum dots. The optical limiting behaviour of the Pc complexes and conjugates are assessed using the open aperture Z–scan technique at laser excitation wavelength of 532 nm with 10 ns pulse. Pcs complexes showed good nonlinear optical response with higher nonlinear absorption coefficient. The conjugates afforded higher nonlinear absorption coefficient than Pc complexes alone.
- Full Text:
- Date Issued: 2019
- Authors: Mgidlana, Sithi
- Date: 2019
- Subjects: Nonlinear optics , Quantum dots , Phthalocyanines , Zinc
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/97152 , vital:31404
- Description: Various characterization techniques have been used to characterize the synthesized asymmetrical zinc phthalocyanines (ZnPc) derivatives. Techniques include Ultraviolet-visible (UV-vis) spectrophotometry, matrix assisted laser desorption time of flight mass spectrometry (MALD-TOF MS), proton nuclear magnetic resonance (1H-NMR), elemental analysis and Fourier-transform infra-red spectroscopy (FT-IR). The complexes are covalently linked to core/shell and core/shell/shell semiconductor quantum dots (SQDs) via amide bond formation. Photophysical properties of complexes improved in the presence of semiconductor quantum dots (SQDs). SQDs contain cadmium/telluride (CdTe) as core, coated in the first shell with zinc selenide (ZnSe) or zinc sulfide (ZnS) and with zinc oxide (ZnO) in second shell. The photophysical properties of the phthalocyanine (Pc) complexes and their conjugates with SQDs are investigated in solution. Triplet quantum yields of complexes improved in the presence of semiconductor quantum dots. The optical limiting behaviour of the Pc complexes and conjugates are assessed using the open aperture Z–scan technique at laser excitation wavelength of 532 nm with 10 ns pulse. Pcs complexes showed good nonlinear optical response with higher nonlinear absorption coefficient. The conjugates afforded higher nonlinear absorption coefficient than Pc complexes alone.
- Full Text:
- Date Issued: 2019
Effects of nitrogen and phosphorus rates, leaf age and drying technique on growth and nutritional contents of Moringa oleifera
- Authors: Sokombela, Asanda
- Date: 2019
- Subjects: Moringa
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/12117 , vital:39153
- Description: In general, smallholder farmers tend to use large quantities of soil nutrients without replenishment with manure, compost and fertiliser in order to replace the soil nutrients that are lost. Considering the low soil fertility status of soils in South Africa, this study was conducted with an aim of developing agronomic protocols for the production of moringa in order to fight malnutrition and produce supplement feed for animals. Moringa contains various nutrients and chemicals, such as, protein, which are sensitive to heat, and there is therefore a need to establish the best drying method which does not affect the nutrient content and quality of moringa. Physiological age is probably the most important factor influencing the mineral composition of plants. The relationship between nutrient concentration and leaf age is not fundamentally altered by crop load, soil type, climate or cultural practices, such that nutrients may be displaced vertically by the level of supply. Leaf age is one of the factors that affect the distribution of nutrients in the leaves of plants. Therefore, studies were conducted to evaluate the response of Moringa oleifera leaf yield and nutritional content to nitrogen and phosphorus rates, leaf age/position, and drying techniques. The experiments were conducted at the University of Fort Hare Alice Campus research farm. The experiment that investigated the effect of N and P rates was carried out in potted plants filled fine silica sand. The seeds were sown on black polythene bags, filled with a mixture of soil, pine bark and goat manure (at 1:2:1 ratio, respectively) at a depth of 2 cm. At 30 cm height, the seedlings were transplanted into 5-liter pots, which were filled with building sand as growing media. The experiment was a 4 x 3 factorial, with four levels of nitrogen (100, 200, 300 and 400 kg/ha) and three levels of phosphorus (40, 80 and 120 kg/ha). The experiment was laid-out as randomised complete block design (RCBD), with 12 treatment combinations replicated four times. All experimental units received the same amount of potassium (731kg/ha) in the form of KCl. Nitrogen treatments which were applied were split in two equal applications, at transplanting stage and 4 weeks later. The phosphorus rates were applied as superphosphate once mixed to the potting media before transplanting. Growth parameters such as plant height, root collar diameters, number of leaves, dry and fresh leaf masses were measured. Powdered dry leaves were also analysed to determine iv metabolite such as crude protein, ash, moisture, fat and fibre constituents (ADF and NDF). Micro and macro mineral nutrients were also determined using atomic absorption spectroscopy technique (AAS). The results showed that growth of moringa significantly increased through the application of fertilizer. The results from the (ANOVA) table showed no interaction between phosphorus and nitrogen on growth parameters such as number of leaves, stem width and plant height of moringa in this study. Nitrogen levels, as treatments, have significantly affected the growth of the plant. However, phosphorus, as a treatment, showed no significant difference on growth parameteres of Moringa. It was evident that trees that received the highest nitrogen rate (400 kg/ha) resulted in thickest stem girth, tallest trees and the highest number of leaves. The overall results revealed that moringa leaf powder contained appreciable amount nutrients. Crude protein was notably high (38.23%) in the trees that received the highest phosphorus and nitrogen combination treatment (400 kg/ha N x 120 kg/ha P treatment combination. The highest ash content of (14.17%) and (14.21%) were recorded from the trees that received treatment combination of 200 kg/ha N x 80 kg/ha P and 100 kg/ha N x 120 kg/ha P, and significantly differed from the rest of the other treatment combinations. The fat content steadily increased with an increase in nitrogen rate. Increase in P level resulted in a decrease in fat content. The overall findings confirm that combinations of the higher levels of N and P result in a decrease in fat content. The variation in the metabolite nutrients and mineral nutrient values may be due to differences in age of the leaves or positions and possibly due to different stages of maturity of leaves. The second experiment was conducted in the field at the research farm. The aim was to evaluate the effect of leaf age/position on nutrient distribution of moringa. The seeds were sown in black polythene bags at a depth of 2 cm. The bags were filled with a mixture of 1 part of soil, 2 parts of pine bark: and 1 part of manure. The seedlings were transplanted into the field when they reached the height of 30cm. Plants were laid-out as a CRD design with 5 replications. Three leaf positions, upper, middle and basal leaves, were used as treatments to determine the nutrient variation within a parent plant. Leaves of each position were harvested separately, washed with tap water and air-dried in shade condition. The dried leaves were analysed for metabolite and mineral content. The Association of official agricultural chemists (AOAC) methods were used to determine protein, fat, fibre, and ash. Atomic absorption spectrometry (AAS) technique was used to analyse minerals v such as iron, zinc, copper, manganese and macro-minerals such as calcium, magnesium, potassium, sodium and phosphorus. The study showed that ash, fat, NDF and ADF contents were significantly higher in the basal leaves as compared to that in the middle and upper leaves. However, moisture and crude protein contents were significantly higher in the upper leaves compared to both in the middle and basal leaves. The basal leaves accumulated significantly higher Ca, Mg, K, Na and Mn contents. The concentrations of P, Zn, Cu and Fe were, on the other hand, significantly higher in the younger leaves compared to that in the middle and basal leaves. It can be concluded that the upper (younger) leaves contain enough nutrition to satisfy the dietary requirements. In addition, these leaves have low levels of oxalate, cyanide and nitrate, which can cause health problems. To determine the effect of drying method on nutrition content of moringa leaves, three drying methods (shade-, sun- and oven-drying methods) were applied as treatments. Moringa plants were grown at the Research Farm of the University of Fort Hare, Alice campus. The leaves were harvested, thoroughly mixed, washed with tap water and put into 15 brown paper bags. The 15 samples (paper bags) were distributed to the three drying treatments (each treatment replicated 5 times). The dried samples were analysed for metabolites and mineral content as per AOAC and AAS methods, respectively. The study showed that shade-drying method retained higher metabolites. Thus, this study recommends the shade-drying method in preserving moringa leaves.
- Full Text:
- Date Issued: 2019
- Authors: Sokombela, Asanda
- Date: 2019
- Subjects: Moringa
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/12117 , vital:39153
- Description: In general, smallholder farmers tend to use large quantities of soil nutrients without replenishment with manure, compost and fertiliser in order to replace the soil nutrients that are lost. Considering the low soil fertility status of soils in South Africa, this study was conducted with an aim of developing agronomic protocols for the production of moringa in order to fight malnutrition and produce supplement feed for animals. Moringa contains various nutrients and chemicals, such as, protein, which are sensitive to heat, and there is therefore a need to establish the best drying method which does not affect the nutrient content and quality of moringa. Physiological age is probably the most important factor influencing the mineral composition of plants. The relationship between nutrient concentration and leaf age is not fundamentally altered by crop load, soil type, climate or cultural practices, such that nutrients may be displaced vertically by the level of supply. Leaf age is one of the factors that affect the distribution of nutrients in the leaves of plants. Therefore, studies were conducted to evaluate the response of Moringa oleifera leaf yield and nutritional content to nitrogen and phosphorus rates, leaf age/position, and drying techniques. The experiments were conducted at the University of Fort Hare Alice Campus research farm. The experiment that investigated the effect of N and P rates was carried out in potted plants filled fine silica sand. The seeds were sown on black polythene bags, filled with a mixture of soil, pine bark and goat manure (at 1:2:1 ratio, respectively) at a depth of 2 cm. At 30 cm height, the seedlings were transplanted into 5-liter pots, which were filled with building sand as growing media. The experiment was a 4 x 3 factorial, with four levels of nitrogen (100, 200, 300 and 400 kg/ha) and three levels of phosphorus (40, 80 and 120 kg/ha). The experiment was laid-out as randomised complete block design (RCBD), with 12 treatment combinations replicated four times. All experimental units received the same amount of potassium (731kg/ha) in the form of KCl. Nitrogen treatments which were applied were split in two equal applications, at transplanting stage and 4 weeks later. The phosphorus rates were applied as superphosphate once mixed to the potting media before transplanting. Growth parameters such as plant height, root collar diameters, number of leaves, dry and fresh leaf masses were measured. Powdered dry leaves were also analysed to determine iv metabolite such as crude protein, ash, moisture, fat and fibre constituents (ADF and NDF). Micro and macro mineral nutrients were also determined using atomic absorption spectroscopy technique (AAS). The results showed that growth of moringa significantly increased through the application of fertilizer. The results from the (ANOVA) table showed no interaction between phosphorus and nitrogen on growth parameters such as number of leaves, stem width and plant height of moringa in this study. Nitrogen levels, as treatments, have significantly affected the growth of the plant. However, phosphorus, as a treatment, showed no significant difference on growth parameteres of Moringa. It was evident that trees that received the highest nitrogen rate (400 kg/ha) resulted in thickest stem girth, tallest trees and the highest number of leaves. The overall results revealed that moringa leaf powder contained appreciable amount nutrients. Crude protein was notably high (38.23%) in the trees that received the highest phosphorus and nitrogen combination treatment (400 kg/ha N x 120 kg/ha P treatment combination. The highest ash content of (14.17%) and (14.21%) were recorded from the trees that received treatment combination of 200 kg/ha N x 80 kg/ha P and 100 kg/ha N x 120 kg/ha P, and significantly differed from the rest of the other treatment combinations. The fat content steadily increased with an increase in nitrogen rate. Increase in P level resulted in a decrease in fat content. The overall findings confirm that combinations of the higher levels of N and P result in a decrease in fat content. The variation in the metabolite nutrients and mineral nutrient values may be due to differences in age of the leaves or positions and possibly due to different stages of maturity of leaves. The second experiment was conducted in the field at the research farm. The aim was to evaluate the effect of leaf age/position on nutrient distribution of moringa. The seeds were sown in black polythene bags at a depth of 2 cm. The bags were filled with a mixture of 1 part of soil, 2 parts of pine bark: and 1 part of manure. The seedlings were transplanted into the field when they reached the height of 30cm. Plants were laid-out as a CRD design with 5 replications. Three leaf positions, upper, middle and basal leaves, were used as treatments to determine the nutrient variation within a parent plant. Leaves of each position were harvested separately, washed with tap water and air-dried in shade condition. The dried leaves were analysed for metabolite and mineral content. The Association of official agricultural chemists (AOAC) methods were used to determine protein, fat, fibre, and ash. Atomic absorption spectrometry (AAS) technique was used to analyse minerals v such as iron, zinc, copper, manganese and macro-minerals such as calcium, magnesium, potassium, sodium and phosphorus. The study showed that ash, fat, NDF and ADF contents were significantly higher in the basal leaves as compared to that in the middle and upper leaves. However, moisture and crude protein contents were significantly higher in the upper leaves compared to both in the middle and basal leaves. The basal leaves accumulated significantly higher Ca, Mg, K, Na and Mn contents. The concentrations of P, Zn, Cu and Fe were, on the other hand, significantly higher in the younger leaves compared to that in the middle and basal leaves. It can be concluded that the upper (younger) leaves contain enough nutrition to satisfy the dietary requirements. In addition, these leaves have low levels of oxalate, cyanide and nitrate, which can cause health problems. To determine the effect of drying method on nutrition content of moringa leaves, three drying methods (shade-, sun- and oven-drying methods) were applied as treatments. Moringa plants were grown at the Research Farm of the University of Fort Hare, Alice campus. The leaves were harvested, thoroughly mixed, washed with tap water and put into 15 brown paper bags. The 15 samples (paper bags) were distributed to the three drying treatments (each treatment replicated 5 times). The dried samples were analysed for metabolites and mineral content as per AOAC and AAS methods, respectively. The study showed that shade-drying method retained higher metabolites. Thus, this study recommends the shade-drying method in preserving moringa leaves.
- Full Text:
- Date Issued: 2019
Environmental drivers of the composition and distribution of larval fish assemblages off the south coast of South Africa
- Authors: Trassierra, Jaqueline Anne
- Date: 2019
- Subjects: Fishes -- Larvae -- South Africa -- Eastern Cape , Fishes -- Larvae -- Migration -- South Africa -- Eastern Cape , Fishes -- Larvae -- Dispersal -- South Africa -- Eastern Cape
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/68154 , vital:29207
- Description: The species composition, distribution and patterns of vertical migration of larval fish assemblages were investigated in March and in September 2013 within two adjacent log spiral bays, Algoa Bay and St Francis Bay, on the south coast of South Africa. Fish larvae were collected by means of a boat towed bongo net (57 cm diameter; mesh aperture 500 μm). An onshore (2 km) and an offshore (3 km) station were each sampled twice during the daytime (06:00 – 18:00) with two horizontal tows: near the surface (0.5 m) and close to the bottom (12 m). Tows were repeated at night (18:00 – 23:00) for onshore sites. Larval catches included 16 fish families and 40 species. A multivariate analysis indicated that the species composition was significantly different between Algoa Bay and St Francis Bay, with Engraulidae, Blenniidae, Sparidae, Soleidae and Cynoglossidae making important contributions to the larval fish catch in Algoa Bay, while Blenniidae, Engraulidae, Tripterygiidae, Sparidae and Gobiesocidae contributed significantly in St Francis Bay. Differences in assemblage composition were noted between the Spring (September–October) and Autumn (March-April) months. The species composition of larval fish assemblages was related to wind speed, wave height, cloud cover, sea water temperature, depth, average current speed and direction. Wind speed, wave height, temperature and depth significantly contributed to the variation in larval fish densities. Abundances of larval fishes were greater offshore than onshore, larvae from pelagic eggs dominated catches offshore, while larvae from demersal eggs dominated onshore catches. Habitat structure strongly influenced the composition of larval fishes between the bays and abundances were significantly greater at night than during the day. Most larval fishes displayed a reverse diel vertical migration pattern and were most influenced by predators, wind speed and cloud cover. This study shows that larval fish assemblages are highly complex and patchy. Spawning mode, individual species behaviour, diel vertical migration, current structure, depth, temperature, wind speed, cloud cover and type of habitat substratum all influence larval fish composition and distribution in the nearshore waters of South Africa.
- Full Text:
- Date Issued: 2019
- Authors: Trassierra, Jaqueline Anne
- Date: 2019
- Subjects: Fishes -- Larvae -- South Africa -- Eastern Cape , Fishes -- Larvae -- Migration -- South Africa -- Eastern Cape , Fishes -- Larvae -- Dispersal -- South Africa -- Eastern Cape
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/68154 , vital:29207
- Description: The species composition, distribution and patterns of vertical migration of larval fish assemblages were investigated in March and in September 2013 within two adjacent log spiral bays, Algoa Bay and St Francis Bay, on the south coast of South Africa. Fish larvae were collected by means of a boat towed bongo net (57 cm diameter; mesh aperture 500 μm). An onshore (2 km) and an offshore (3 km) station were each sampled twice during the daytime (06:00 – 18:00) with two horizontal tows: near the surface (0.5 m) and close to the bottom (12 m). Tows were repeated at night (18:00 – 23:00) for onshore sites. Larval catches included 16 fish families and 40 species. A multivariate analysis indicated that the species composition was significantly different between Algoa Bay and St Francis Bay, with Engraulidae, Blenniidae, Sparidae, Soleidae and Cynoglossidae making important contributions to the larval fish catch in Algoa Bay, while Blenniidae, Engraulidae, Tripterygiidae, Sparidae and Gobiesocidae contributed significantly in St Francis Bay. Differences in assemblage composition were noted between the Spring (September–October) and Autumn (March-April) months. The species composition of larval fish assemblages was related to wind speed, wave height, cloud cover, sea water temperature, depth, average current speed and direction. Wind speed, wave height, temperature and depth significantly contributed to the variation in larval fish densities. Abundances of larval fishes were greater offshore than onshore, larvae from pelagic eggs dominated catches offshore, while larvae from demersal eggs dominated onshore catches. Habitat structure strongly influenced the composition of larval fishes between the bays and abundances were significantly greater at night than during the day. Most larval fishes displayed a reverse diel vertical migration pattern and were most influenced by predators, wind speed and cloud cover. This study shows that larval fish assemblages are highly complex and patchy. Spawning mode, individual species behaviour, diel vertical migration, current structure, depth, temperature, wind speed, cloud cover and type of habitat substratum all influence larval fish composition and distribution in the nearshore waters of South Africa.
- Full Text:
- Date Issued: 2019
Epibenthic biodiversity, habitat characterisation and anthropogenic pressure mapping of unconsolidated sediment habitats in Algoa Bay, South Africa
- Authors: Truter, Hannah Jessie
- Date: 2019
- Subjects: Aquatic ecology , Marine sciences Conservation biology Ecology
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/43812 , vital:37049
- Description: Implementation of an ecosystem-based management approach for marine systems requires a comprehensive understanding of the biophysical marine environment and the cumulative human impacts at different spatio-temporal scales. In Algoa Bay, South Africa, this study describes the epibenthic communities occurring in unconsolidated marine habitats. It further investigates the potential abiotic factors that influence their distribution and abundance, compares epibenthic communities with existing habitat information and evaluates the protection status of the marine environment in the Bay. Seabed imagery, covering a total area of 171.4m², and sediment samples were collected from 13 stations from which 106 epibenthic species were identified. Multivariate analyses revealed two statistically distinct communities that did not align with the Algoa Bay benthic habitat types defined in the current National Biodiversity Assessment (NBA, 2012). Further assessment indicated that community differences were driven by the presence of rock substrate. A range of abiotic factors were tested against the epibenthic communities to explore patterns and identify potential drivers. The combination of abiotic factors depth, mean grain size, mean bottom temperature and mean bottom current explained 55% fitted variation in epibenthic data. The degree of long-term variability in several of these parameters were likewise identified as explanatory variables, including bottom temperature, current speed and dissolved oxygen. The link between abiotic factors and the epibenthic communities observed indicate that these variables can act as surrogates for habitat mapping in the future. The existing and proposed Marine Protected Area (MPA) in conjunction with the NBA 2012 habitat types does well in protecting the majority of habitats in the Bay, however there remain habitats that lack protection. Utilising the benthic communities and potential drivers identified in this study, the proposed MPA boundary delineations should be somewhat altered to include missing habitat types.
- Full Text:
- Date Issued: 2019
- Authors: Truter, Hannah Jessie
- Date: 2019
- Subjects: Aquatic ecology , Marine sciences Conservation biology Ecology
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/43812 , vital:37049
- Description: Implementation of an ecosystem-based management approach for marine systems requires a comprehensive understanding of the biophysical marine environment and the cumulative human impacts at different spatio-temporal scales. In Algoa Bay, South Africa, this study describes the epibenthic communities occurring in unconsolidated marine habitats. It further investigates the potential abiotic factors that influence their distribution and abundance, compares epibenthic communities with existing habitat information and evaluates the protection status of the marine environment in the Bay. Seabed imagery, covering a total area of 171.4m², and sediment samples were collected from 13 stations from which 106 epibenthic species were identified. Multivariate analyses revealed two statistically distinct communities that did not align with the Algoa Bay benthic habitat types defined in the current National Biodiversity Assessment (NBA, 2012). Further assessment indicated that community differences were driven by the presence of rock substrate. A range of abiotic factors were tested against the epibenthic communities to explore patterns and identify potential drivers. The combination of abiotic factors depth, mean grain size, mean bottom temperature and mean bottom current explained 55% fitted variation in epibenthic data. The degree of long-term variability in several of these parameters were likewise identified as explanatory variables, including bottom temperature, current speed and dissolved oxygen. The link between abiotic factors and the epibenthic communities observed indicate that these variables can act as surrogates for habitat mapping in the future. The existing and proposed Marine Protected Area (MPA) in conjunction with the NBA 2012 habitat types does well in protecting the majority of habitats in the Bay, however there remain habitats that lack protection. Utilising the benthic communities and potential drivers identified in this study, the proposed MPA boundary delineations should be somewhat altered to include missing habitat types.
- Full Text:
- Date Issued: 2019
Evaluation of some Eastern beach bacteria for high activity laccase production
- Authors: Ntlatywa, Nonkolisi
- Date: 2019
- Subjects: Laccase
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/${Handle} , vital:40025
- Description: Laccases are multi-copper oxidases endowed with huge applicability and versatility. They are found in nature, fungi, insects, archaea and bacteria; however, there is a dearth of information regarding bacterial laccases. Thus, the search for novel bacterial laccases becomes crucial. Therefore, the broad aim of the study was the assessment of some bacteria on marine water isolated from Eastern beach for the production of high activity laccases. In this study, marine water samples obtained from the Eastern beach in East London were investigated for bacterial laccase-producing isolates. The samples were analysed using different microbiological media and varied substrates via qualitative methods. The isolates were tested for laccase activity using ABTS (2, 2’- azinobis 3-ethyl-benzothiazoline-6-sulphonate), as a substrate. The present study also dealt with the optimization of culture conditions for laccase production and characterization of laccase. A total of forty-eight (48) bacterial isolates were obtained, which were identified based on their morphological and cultural characteristics. Many of the isolates (42) were categorized as Gram positive and a few (6) as Gram negative upon subjecting to microscopic examination. Result showed that on average, laccase activity ranged from 5.33 – 6.5 U/L as demonstrated by three bacterial isolates (Bacillus sp. NFN1, Bacillus sp. NLN and Escherichia coli LO). Laccase production was highly enhanced by the supplementation with vanillic acid by Bacillus sp. strain NLN with yield of 15.83 ± 0.7 U/L, Bacillus sp. strain NFN1, 12.72 ± 0.21 U/L. However, Escherichia coli LO demonstrated enhanced laccase production with guaiacol (12.84 ± 1.12 U/L) followed by ferrulic acid (12.8 ± 0.36 U/L). While optimization results revealed that the laccase yield was enhanced under the following conditions: 72 h of incubation, pH 5, 0.5mg/L lactose as a carbon source, 1. 5 mg/L yeast extract as a nitrogen source as well as in the presence of corn stover and mandarin peels as lignocellulosic substrates depending on the bacteria.The time course and stability of laccase against temperature, pH, metal ions and organic solvents were investigated using previously described methods. In the findings obtained for the time course, the laccase activity ranged from 22.2- 23.98 U/L. In general, the optimum parameters for the optimum laccase activities depended on the bacterial isolates and the time of incubation and included 80 °C and 70 °C, pH 3 and 9 while, the thermal stability was over 100%. Overall, NaCl, ZnSO4, NiCl and 20% acetone demonstrated positive effect on the residual activity of the laccases of all the bacterial isolates (Bacillus sp. NFN1, Bacillus v sp. NLN and Escherichia coli LO). In conclusion the findings of the study, suggest that marine water of the Eastern beach is a rich source for laccase producing bacteria with potential of high activity.
- Full Text:
- Date Issued: 2019
- Authors: Ntlatywa, Nonkolisi
- Date: 2019
- Subjects: Laccase
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/${Handle} , vital:40025
- Description: Laccases are multi-copper oxidases endowed with huge applicability and versatility. They are found in nature, fungi, insects, archaea and bacteria; however, there is a dearth of information regarding bacterial laccases. Thus, the search for novel bacterial laccases becomes crucial. Therefore, the broad aim of the study was the assessment of some bacteria on marine water isolated from Eastern beach for the production of high activity laccases. In this study, marine water samples obtained from the Eastern beach in East London were investigated for bacterial laccase-producing isolates. The samples were analysed using different microbiological media and varied substrates via qualitative methods. The isolates were tested for laccase activity using ABTS (2, 2’- azinobis 3-ethyl-benzothiazoline-6-sulphonate), as a substrate. The present study also dealt with the optimization of culture conditions for laccase production and characterization of laccase. A total of forty-eight (48) bacterial isolates were obtained, which were identified based on their morphological and cultural characteristics. Many of the isolates (42) were categorized as Gram positive and a few (6) as Gram negative upon subjecting to microscopic examination. Result showed that on average, laccase activity ranged from 5.33 – 6.5 U/L as demonstrated by three bacterial isolates (Bacillus sp. NFN1, Bacillus sp. NLN and Escherichia coli LO). Laccase production was highly enhanced by the supplementation with vanillic acid by Bacillus sp. strain NLN with yield of 15.83 ± 0.7 U/L, Bacillus sp. strain NFN1, 12.72 ± 0.21 U/L. However, Escherichia coli LO demonstrated enhanced laccase production with guaiacol (12.84 ± 1.12 U/L) followed by ferrulic acid (12.8 ± 0.36 U/L). While optimization results revealed that the laccase yield was enhanced under the following conditions: 72 h of incubation, pH 5, 0.5mg/L lactose as a carbon source, 1. 5 mg/L yeast extract as a nitrogen source as well as in the presence of corn stover and mandarin peels as lignocellulosic substrates depending on the bacteria.The time course and stability of laccase against temperature, pH, metal ions and organic solvents were investigated using previously described methods. In the findings obtained for the time course, the laccase activity ranged from 22.2- 23.98 U/L. In general, the optimum parameters for the optimum laccase activities depended on the bacterial isolates and the time of incubation and included 80 °C and 70 °C, pH 3 and 9 while, the thermal stability was over 100%. Overall, NaCl, ZnSO4, NiCl and 20% acetone demonstrated positive effect on the residual activity of the laccases of all the bacterial isolates (Bacillus sp. NFN1, Bacillus v sp. NLN and Escherichia coli LO). In conclusion the findings of the study, suggest that marine water of the Eastern beach is a rich source for laccase producing bacteria with potential of high activity.
- Full Text:
- Date Issued: 2019
Examining the influence of extrinsic factors on herbivore assemblage composition in a southern African savanna
- Young, Cyanne, Smithwick, Erica
- Authors: Young, Cyanne , Smithwick, Erica
- Date: 2019
- Subjects: Herbivores -- Ecology , Ungulates -- Ecology Mammal populations -- South Africa
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/44334 , vital:37154
- Description: Globally, herbivore populations have become threatened by the effects of a growing human population. This has resulted in over-hunting, habitat encroachment, fragmentation and degradation and competition leading to the replacement of wild herbivore populations with livestock. African herbivores, in particular, have experienced a severe decline in abundance and distribution, and vast conservation management efforts are underway to preserve herbivore populations and promote their growth. However, conservation management efforts sometimes have unintended consequences on the abundance and distribution of herbivores as a result of the complex interactions they have with their ecosystem. The aim of this study was to: 1) examine the drivers of herbivore distribution at the Satara section of Kruger National Park (hereafter ‘KNP’) at a landscape scale, to determine assemblage composition under the constraints of different environmental attributes; and 2) examine feeding selection by grazing herbivores of different morpho-physiological characteristics at a patch scale of differing environmental attributes. To achieve this, thirty-six experimental plots were established at three distances from water at Satara, and manipulated by mowing, mowing and fertilising or fertilising only over the course of a year, to account for seasonal effects. Each plot was fitted with a camera trap which ran for five weeks in each sampling season, and the data collected focussed on eleven herbivore species of differing body size, digestive strategy and feeding preference. These species were selected due to their abundance at Satara, and their representation of a variety of morpho-physiological characteristics. To achieve both aims, generalised linear models were used to determine the probability of occurrence of a functional type (for aim 1- Chapter 3), and individual species (for aim 2- Chapter 4). To construct the generalised linear models for Chapter 3, hypotheses about species’ relative distribution and abundance were developed through a literature review of forage and water availability constraints on feeding preference and body size of herbivore biomass at a landscape scale. I expected strong seasonal relationships between vegetation biomass and quality, and biomass of water-dependent herbivores with increasing distance to water. My analyses of herbivore distribution across the region confirmed broad-scale descriptions of interactions between forage requirements and water availability, across a set of species which differ in functional traits, at a landscape scale. In Chapter 4, trends in probability of occurrence by seven grazing herbivore species were determined using camera trap data at a patch scale. My results showed that season was a major determinant of species distribution, especially those which are not obligate grazers or feed exclusively in the 0.5 km to 2 km zone from water. I found that most selective feeding occurred in the late wet season when water would be more evenly distributed across the landscape and forage resources close to water would have had the chance to recover from depletion as a result of dry season use. The collective efforts of this study show that distance from water was a critical determinant of herbivore distribution across the landscape, and that forage utilisation by small mixed-feeders and large grazers in particular, altered significantly across seasons. This has important implications for the provision of artificial water, given that foraging pressure is increased closer to water. Landscape heterogeneity needs to be maintained through efficient water provision, to allow areas of forage reserve to maintained in the dry season, where species are constrained by water availability and forage quality and quantity.
- Full Text:
- Date Issued: 2019
- Authors: Young, Cyanne , Smithwick, Erica
- Date: 2019
- Subjects: Herbivores -- Ecology , Ungulates -- Ecology Mammal populations -- South Africa
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/44334 , vital:37154
- Description: Globally, herbivore populations have become threatened by the effects of a growing human population. This has resulted in over-hunting, habitat encroachment, fragmentation and degradation and competition leading to the replacement of wild herbivore populations with livestock. African herbivores, in particular, have experienced a severe decline in abundance and distribution, and vast conservation management efforts are underway to preserve herbivore populations and promote their growth. However, conservation management efforts sometimes have unintended consequences on the abundance and distribution of herbivores as a result of the complex interactions they have with their ecosystem. The aim of this study was to: 1) examine the drivers of herbivore distribution at the Satara section of Kruger National Park (hereafter ‘KNP’) at a landscape scale, to determine assemblage composition under the constraints of different environmental attributes; and 2) examine feeding selection by grazing herbivores of different morpho-physiological characteristics at a patch scale of differing environmental attributes. To achieve this, thirty-six experimental plots were established at three distances from water at Satara, and manipulated by mowing, mowing and fertilising or fertilising only over the course of a year, to account for seasonal effects. Each plot was fitted with a camera trap which ran for five weeks in each sampling season, and the data collected focussed on eleven herbivore species of differing body size, digestive strategy and feeding preference. These species were selected due to their abundance at Satara, and their representation of a variety of morpho-physiological characteristics. To achieve both aims, generalised linear models were used to determine the probability of occurrence of a functional type (for aim 1- Chapter 3), and individual species (for aim 2- Chapter 4). To construct the generalised linear models for Chapter 3, hypotheses about species’ relative distribution and abundance were developed through a literature review of forage and water availability constraints on feeding preference and body size of herbivore biomass at a landscape scale. I expected strong seasonal relationships between vegetation biomass and quality, and biomass of water-dependent herbivores with increasing distance to water. My analyses of herbivore distribution across the region confirmed broad-scale descriptions of interactions between forage requirements and water availability, across a set of species which differ in functional traits, at a landscape scale. In Chapter 4, trends in probability of occurrence by seven grazing herbivore species were determined using camera trap data at a patch scale. My results showed that season was a major determinant of species distribution, especially those which are not obligate grazers or feed exclusively in the 0.5 km to 2 km zone from water. I found that most selective feeding occurred in the late wet season when water would be more evenly distributed across the landscape and forage resources close to water would have had the chance to recover from depletion as a result of dry season use. The collective efforts of this study show that distance from water was a critical determinant of herbivore distribution across the landscape, and that forage utilisation by small mixed-feeders and large grazers in particular, altered significantly across seasons. This has important implications for the provision of artificial water, given that foraging pressure is increased closer to water. Landscape heterogeneity needs to be maintained through efficient water provision, to allow areas of forage reserve to maintained in the dry season, where species are constrained by water availability and forage quality and quantity.
- Full Text:
- Date Issued: 2019
Exploration of challenges in bringing traditional medicine into SA’s healthcare system, using medicinal plants for treatment of waterborne diarrhoeal diseases as a case study
- Authors: Keche, Priscilla
- Date: 2019
- Subjects: Traditional medicine -- South Africa , Waterborne infection -- South Africa , Diarrhea -- South Africa , Healers -- South Africa , Medical care -- Quality control , Medicinal plants -- South Africa , Diarrhea in children -- South Africa , World Health Organization
- Language: English
- Type: thesis , text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/118035 , vital:34588
- Description: Thesis (MSc)--Rhodes University, Faculty of Science, Biotechnology Innovation Centre (RUBIC), 2019.
- Full Text:
- Date Issued: 2019
- Authors: Keche, Priscilla
- Date: 2019
- Subjects: Traditional medicine -- South Africa , Waterborne infection -- South Africa , Diarrhea -- South Africa , Healers -- South Africa , Medical care -- Quality control , Medicinal plants -- South Africa , Diarrhea in children -- South Africa , World Health Organization
- Language: English
- Type: thesis , text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/118035 , vital:34588
- Description: Thesis (MSc)--Rhodes University, Faculty of Science, Biotechnology Innovation Centre (RUBIC), 2019.
- Full Text:
- Date Issued: 2019
Exploring the influence of management practice on mesopredator and herbivore occupancy and interactions
- Authors: Winterton, Deborah Jean
- Date: 2019
- Subjects: Herbivores -- Habitat -- Conservation , Herbivores -- Habitat Herbivores -- Conservation Herbivores -- Ecology
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/44368 , vital:37159
- Description: Ecological studies need to consider ecological interactions between species and their environment across trophic levels. This complexity not only makes the study of ecology very challenging but it also means that ecosystems are vulnerable to change as a disruption at any of the levels could result in cascading effects through the hierarchy. This sensitivity to change makes it especially important to understand ecosystem function as this is needed for effective and adaptive conservation management. A key way in which humans drive and change ecosystem function is through land use and associated management practice. A primary objective of many protected areas is restoration of natural function through re-introduction of large ungulates, which are often fenced and small, thus requiring intensive management that can influence ecosystem function. This is true of the West Coast National Park (the park). Small antelope abundance, and associated drivers, in the contractual Postberg section of the park have been a long-term management question. Postberg is a small (1800 ha) fenced and isolated section of the park which was historically used for agriculture (livestock grazing and some cultivation) and later large wild ungulate species were re-introduced and kept at high densities. The perception of a lower abundance of small antelope has been attributed to predation by a mesopredator (Caracal caracal), however interspecific competition and habitat quality is also known to influence species abundance. Therefore, I aimed to explore the potential mechanisms of small antelope occurrence in the region and how this might differ across three sites with different management practices. Using 18 camera traps, I documented the occurrence of small, medium and large ungulates and caracal. My study spanned across three areas that each represented different forms of management practice with regards to fencing, and stocking rates of managed ungulate species. I made use of single-season, single-species occupancy models in R to assess occupancy of small antelope and caracal and I used the abundance induced heterogeneity model to estimate abundance of managed ungulates. I employed the single-season, two-species occupancy model in PRESENCE to explore species co-occurrence and interactions. Activity patterns and temporal overlap between managed ungulates, small antelope and caracal were assessed using the overlap and activity packages in R. Vegetation height appears to be an important driver of common duiker occupancy. Fallow lands were strongly favoured by managed ungulates suggesting that they may be utilising these patches for foraging. Small antelope occupancy was highest outside of the park and caracal detections were lowest outside of the park. The low detection of caracal across all areas, along with the known diet of caracals, suggests that predation is not likely the driver of small antelope occurrence. Our data suggest that co-occurrence between small antelope and managed ungulates is high and that these sympatric species have probably partitioned food resources. I found that small antelope were mostly crepuscular, managed ungulates more diurnal while caracal were primarily nocturnal. There was a high overlap between small antelope and managed ungulate activity while overlap between caracal and small antelope indicates some temporal partitioning. Overall I found some effects of inter-specific interactions at the local scale. There was, however, no consistent pattern across the areas which, in line with literature, suggests large scale ecological trends are difficult to detect at local scales. Due to the heterogeneity of the region’s vegetation, coupled with the small size of the biome and extensive fragmentation by land use, there are few areas which adequately represent the biome to test how large ungulates would utilise the area at a large scale. This has important implications for the management of small protected areas in the region who are mandated to restore these areas to their historical and natural function. The ability of conservation managers to meet objectives of restoring parks to their historical function, and maintaining these systems as such, may be questioned if our protected areas are not meeting the spatial requirements of re-introduced species.
- Full Text:
- Date Issued: 2019
- Authors: Winterton, Deborah Jean
- Date: 2019
- Subjects: Herbivores -- Habitat -- Conservation , Herbivores -- Habitat Herbivores -- Conservation Herbivores -- Ecology
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/44368 , vital:37159
- Description: Ecological studies need to consider ecological interactions between species and their environment across trophic levels. This complexity not only makes the study of ecology very challenging but it also means that ecosystems are vulnerable to change as a disruption at any of the levels could result in cascading effects through the hierarchy. This sensitivity to change makes it especially important to understand ecosystem function as this is needed for effective and adaptive conservation management. A key way in which humans drive and change ecosystem function is through land use and associated management practice. A primary objective of many protected areas is restoration of natural function through re-introduction of large ungulates, which are often fenced and small, thus requiring intensive management that can influence ecosystem function. This is true of the West Coast National Park (the park). Small antelope abundance, and associated drivers, in the contractual Postberg section of the park have been a long-term management question. Postberg is a small (1800 ha) fenced and isolated section of the park which was historically used for agriculture (livestock grazing and some cultivation) and later large wild ungulate species were re-introduced and kept at high densities. The perception of a lower abundance of small antelope has been attributed to predation by a mesopredator (Caracal caracal), however interspecific competition and habitat quality is also known to influence species abundance. Therefore, I aimed to explore the potential mechanisms of small antelope occurrence in the region and how this might differ across three sites with different management practices. Using 18 camera traps, I documented the occurrence of small, medium and large ungulates and caracal. My study spanned across three areas that each represented different forms of management practice with regards to fencing, and stocking rates of managed ungulate species. I made use of single-season, single-species occupancy models in R to assess occupancy of small antelope and caracal and I used the abundance induced heterogeneity model to estimate abundance of managed ungulates. I employed the single-season, two-species occupancy model in PRESENCE to explore species co-occurrence and interactions. Activity patterns and temporal overlap between managed ungulates, small antelope and caracal were assessed using the overlap and activity packages in R. Vegetation height appears to be an important driver of common duiker occupancy. Fallow lands were strongly favoured by managed ungulates suggesting that they may be utilising these patches for foraging. Small antelope occupancy was highest outside of the park and caracal detections were lowest outside of the park. The low detection of caracal across all areas, along with the known diet of caracals, suggests that predation is not likely the driver of small antelope occurrence. Our data suggest that co-occurrence between small antelope and managed ungulates is high and that these sympatric species have probably partitioned food resources. I found that small antelope were mostly crepuscular, managed ungulates more diurnal while caracal were primarily nocturnal. There was a high overlap between small antelope and managed ungulate activity while overlap between caracal and small antelope indicates some temporal partitioning. Overall I found some effects of inter-specific interactions at the local scale. There was, however, no consistent pattern across the areas which, in line with literature, suggests large scale ecological trends are difficult to detect at local scales. Due to the heterogeneity of the region’s vegetation, coupled with the small size of the biome and extensive fragmentation by land use, there are few areas which adequately represent the biome to test how large ungulates would utilise the area at a large scale. This has important implications for the management of small protected areas in the region who are mandated to restore these areas to their historical and natural function. The ability of conservation managers to meet objectives of restoring parks to their historical function, and maintaining these systems as such, may be questioned if our protected areas are not meeting the spatial requirements of re-introduced species.
- Full Text:
- Date Issued: 2019
Extraction, isolation and characterization of oleanolic acid and its analogues from Syzygium aromaticum (cloves) and evaluation of their biological activities
- Authors: khwaza, Vuyolwethu
- Date: 2019
- Subjects: Antibacterial agents
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/19076 , vital:39882
- Description: Pathogenic microorganisms have serious impact on people's lives. Every year, millions of people around the world die of bacterial infections. Resistance to common antibacterial drugs has proven to be a challenging problem in control of bacterial infections. In an attempt to develop an effective and affordable treatment for bacterial infections, oleanolic acid isolated from syzygium aromaticum conjugates incorporating other pharmaceutical scaffolds such as chloroquine derivatives, curcumin, and ergocalciferol etc have been developed. Based on the previous successes of testing combination of antimicrobial drugs and pharmaceutical drugs which appeared to be the promising strategy to overcome treatment failure; a series of hybrid compounds containing oleanolic acid and other pharmaceutical scaffolds were synthesized. 4- Aminoquinoline derivatives were first hybridized with selected organic compounds to form a class of hybrid compounds containing either amide bond or ester bond as a linker between the precursor molecules. Analogues/hybrid compounds can overcome the disadvantages of combination therapy such as drug-drug interaction. The structural effects of this type of conjugation of oleanolic acid and other pharmaceutical scaffolds were characterised by FTIR, Mass Spec and NMR spectroscopy. These compounds were studied along with the monosubstituted oleanolic acid analogues and the organic components in order to compare the effects of the substitution on their biological response.‖ All the synthesized analogues were tested against 11 bacterial strains on both Gram-positive and Gram-negative bacteria. The synthesized compounds showed selectivity and higher activity against Enterococcus faecalis (EF), Klebsiella oxytoca (KO), Escherischia coli (EC), Staphylococcus aureous (SA), Proteus vulgaris (PV) and Bacillus subtilis (BS) with MIC values; ranging between of 1.25 mg/mL to 0.072 mg/mL
- Full Text:
- Date Issued: 2019
- Authors: khwaza, Vuyolwethu
- Date: 2019
- Subjects: Antibacterial agents
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/19076 , vital:39882
- Description: Pathogenic microorganisms have serious impact on people's lives. Every year, millions of people around the world die of bacterial infections. Resistance to common antibacterial drugs has proven to be a challenging problem in control of bacterial infections. In an attempt to develop an effective and affordable treatment for bacterial infections, oleanolic acid isolated from syzygium aromaticum conjugates incorporating other pharmaceutical scaffolds such as chloroquine derivatives, curcumin, and ergocalciferol etc have been developed. Based on the previous successes of testing combination of antimicrobial drugs and pharmaceutical drugs which appeared to be the promising strategy to overcome treatment failure; a series of hybrid compounds containing oleanolic acid and other pharmaceutical scaffolds were synthesized. 4- Aminoquinoline derivatives were first hybridized with selected organic compounds to form a class of hybrid compounds containing either amide bond or ester bond as a linker between the precursor molecules. Analogues/hybrid compounds can overcome the disadvantages of combination therapy such as drug-drug interaction. The structural effects of this type of conjugation of oleanolic acid and other pharmaceutical scaffolds were characterised by FTIR, Mass Spec and NMR spectroscopy. These compounds were studied along with the monosubstituted oleanolic acid analogues and the organic components in order to compare the effects of the substitution on their biological response.‖ All the synthesized analogues were tested against 11 bacterial strains on both Gram-positive and Gram-negative bacteria. The synthesized compounds showed selectivity and higher activity against Enterococcus faecalis (EF), Klebsiella oxytoca (KO), Escherischia coli (EC), Staphylococcus aureous (SA), Proteus vulgaris (PV) and Bacillus subtilis (BS) with MIC values; ranging between of 1.25 mg/mL to 0.072 mg/mL
- Full Text:
- Date Issued: 2019
Factors influencing survivorship of Portulacaria afra cuttings in restoration of degraded lands
- Authors: Sholto-Douglas, Craig
- Date: 2019
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/92230 , vital:30692
- Description: Heavy over-browsing, over-grazing, and over-trampling has caused the degradation of vast areas of subtropical thicket, including parts of the Greater Addo Elephant National Park (GAENP). Severe degradation has resulted in the loss of vegetation cover and the deterioration of soils and soil functioning. The South African Department of Environmental Affairs (DEA), and specifically the Natural Resource Management Programme (NRM), have invested in restoration initiatives such as the Subtropical Thicket Restoration Programme (STRP), which has aimed to demonstrate the logistical and practical feasibility of restoring these degraded thickets by planting en masse Portulacaria afra (spekboom) cuttings. However, the planted P. afra cuttings often succumb to various abiotic and biotic stresses. This project aimed to identify various factors which may influence the survival of P. afra cuttings, including i) soil properties, ii) water (rainfall, infiltration, run-off, etc.), iii) browsing by mega-herbivores, iv) topographical features such as slope and aspect, or v) plant species or communities. Biodiversity and carbon baselines were carried out at 50 plots within the Main Camp, Darlington and Kabouga sections of the GAENP, providing pre-planting vegetation and soil data. The en masse planting of P. afra occurred following the completion of the baselines. Following a minimum rest period of three months after planting, survivorship counts were carried out at all plots. Infiltration rate had a significant positive relationship with planted P. afra survivorship across all of the sites. Positive trends between increases in litter and root biomass and planted P. afra cuttings survivorship in Kabouga were also evident. Magnesium, water-holding capacity and increasing soil clay percentage were found to have a significant negative relationship with planted P. afra survivorship in Kabouga. Aluminium, magnesium, and boron all had significant positive relationships with planted P. afra cuttings survivorship in Addo Main Camp. No soil variables had any significant relationship with survivorship of planted P. afra cuttings at Darlington. The presence of elephants (Loxidonta africana) had a significant negative relationship with planted P. afra survivorship. Plant communities in more ‘moderately’ degraded states, characterized by the retention of canopy dominant species, woody vegetation cover, limited bareground and higher cover of litter had higher survivorship of planted P. afra cuttings than those in more severely degraded states. The grass Panicum maximum, and forb Atriplex semibaccata were associated with positive relationships with planted P. afra survivorship across all sites. Future research should aim to develop matrices which plot likely survivorship percentages with parameters of measurable variables, such as infiltration, severity of soil capping, browsing densities, chemical indicators, and degradation severity, to identify suitable biophysical planting conditions for P. afra cuttings. Further research should assess whether the presence of elephants, in high densities, can reduce the economic feasibility of the en masse plantings of P. afra cuttings in attempts to restore degraded landscapes.
- Full Text:
- Date Issued: 2019
- Authors: Sholto-Douglas, Craig
- Date: 2019
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/92230 , vital:30692
- Description: Heavy over-browsing, over-grazing, and over-trampling has caused the degradation of vast areas of subtropical thicket, including parts of the Greater Addo Elephant National Park (GAENP). Severe degradation has resulted in the loss of vegetation cover and the deterioration of soils and soil functioning. The South African Department of Environmental Affairs (DEA), and specifically the Natural Resource Management Programme (NRM), have invested in restoration initiatives such as the Subtropical Thicket Restoration Programme (STRP), which has aimed to demonstrate the logistical and practical feasibility of restoring these degraded thickets by planting en masse Portulacaria afra (spekboom) cuttings. However, the planted P. afra cuttings often succumb to various abiotic and biotic stresses. This project aimed to identify various factors which may influence the survival of P. afra cuttings, including i) soil properties, ii) water (rainfall, infiltration, run-off, etc.), iii) browsing by mega-herbivores, iv) topographical features such as slope and aspect, or v) plant species or communities. Biodiversity and carbon baselines were carried out at 50 plots within the Main Camp, Darlington and Kabouga sections of the GAENP, providing pre-planting vegetation and soil data. The en masse planting of P. afra occurred following the completion of the baselines. Following a minimum rest period of three months after planting, survivorship counts were carried out at all plots. Infiltration rate had a significant positive relationship with planted P. afra survivorship across all of the sites. Positive trends between increases in litter and root biomass and planted P. afra cuttings survivorship in Kabouga were also evident. Magnesium, water-holding capacity and increasing soil clay percentage were found to have a significant negative relationship with planted P. afra survivorship in Kabouga. Aluminium, magnesium, and boron all had significant positive relationships with planted P. afra cuttings survivorship in Addo Main Camp. No soil variables had any significant relationship with survivorship of planted P. afra cuttings at Darlington. The presence of elephants (Loxidonta africana) had a significant negative relationship with planted P. afra survivorship. Plant communities in more ‘moderately’ degraded states, characterized by the retention of canopy dominant species, woody vegetation cover, limited bareground and higher cover of litter had higher survivorship of planted P. afra cuttings than those in more severely degraded states. The grass Panicum maximum, and forb Atriplex semibaccata were associated with positive relationships with planted P. afra survivorship across all sites. Future research should aim to develop matrices which plot likely survivorship percentages with parameters of measurable variables, such as infiltration, severity of soil capping, browsing densities, chemical indicators, and degradation severity, to identify suitable biophysical planting conditions for P. afra cuttings. Further research should assess whether the presence of elephants, in high densities, can reduce the economic feasibility of the en masse plantings of P. afra cuttings in attempts to restore degraded landscapes.
- Full Text:
- Date Issued: 2019
Finding Frankenflora: investigating hybridisation between local and introduced Protea species at the Van Stadens Wildflower Reserve
- Authors: Macqueen, Timothy
- Date: 2019
- Subjects: Proteaceae -- South Africa , Proteaceae Trees -- South Africa Hybridization
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/40724 , vital:36230
- Description: This study provides an overview of hybridisation and methods to detect it in plants. This is documented in chapter 1. Another aim was to identify hybridisation between local and non-local species in the Proteaceae. This is covered in chapter 2. Finally the third aim was to develop a cheap and rapidly working method of detecting hybrids in this system. This is detailed in chapter 3. The model system for this research is the Van Stadens Wildflower Reserve. In 1984 the reserve manager introduced the non-local species, Protea susannae from Cape Agulhas. This represents a long-distance dispersal of over 600 km. This study is focused on looking at the impacts of the introduction of P. susannae on the local species P. eximia. I use genetic and morphological methods to detect hybrids. Hybridisation and the methods implemented to identify it are reviewed and discussed (Chapter 1). Morphological traits of the two species and direct sequencing of the nuclear ITS and chloroplast regions are compared (Chapter 2). This includes notes on the detection of hybrids and the potential for cryptic hybrids. I also focussed on testing of two rapid and low-cost techniques for detecting hybrids using Species Specific Primers (SSP’s) and High Resolution Melt (HRM). The Sanger sequencing method and ITS detected cryptic hybrids. The HRM was able to detect hybrids when comparing F1 putative hybrids to a ‘pure’ parent populations, but was not able to detect between a ‘pure’ parent population and F3 or F4 putative hybrids in the Van Stadens system. The SSP’s had the highest hybrid detection rate. Genetic methods, specifically the SSP’s were found to work well to detect hybrids in an environment where a hybrid swarm has taken place. The maps produced from this thesis will be useful for determining the distribution of hybrids in the reserve and for other similar hybrid systems. Long distance dispersal of Cape species may lead to the loss of genetic diversity or species as local and non-local hybridisation may occur, similar to the hybridisation between P. susannae and P. eximia in the Van Stadens Wild Flower Reserve. Genetic methods were shown to be superior to morphological hybrid detection (i.e. Sanger sequencing and SSP’s). The SSP approach tested here, which is dependent on the presence of fixed species specific SNP’s (Single Nucleotide Polymorphisms), could offer a rapid and effective method to explore hybridisation through space and time.
- Full Text:
- Date Issued: 2019
- Authors: Macqueen, Timothy
- Date: 2019
- Subjects: Proteaceae -- South Africa , Proteaceae Trees -- South Africa Hybridization
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/40724 , vital:36230
- Description: This study provides an overview of hybridisation and methods to detect it in plants. This is documented in chapter 1. Another aim was to identify hybridisation between local and non-local species in the Proteaceae. This is covered in chapter 2. Finally the third aim was to develop a cheap and rapidly working method of detecting hybrids in this system. This is detailed in chapter 3. The model system for this research is the Van Stadens Wildflower Reserve. In 1984 the reserve manager introduced the non-local species, Protea susannae from Cape Agulhas. This represents a long-distance dispersal of over 600 km. This study is focused on looking at the impacts of the introduction of P. susannae on the local species P. eximia. I use genetic and morphological methods to detect hybrids. Hybridisation and the methods implemented to identify it are reviewed and discussed (Chapter 1). Morphological traits of the two species and direct sequencing of the nuclear ITS and chloroplast regions are compared (Chapter 2). This includes notes on the detection of hybrids and the potential for cryptic hybrids. I also focussed on testing of two rapid and low-cost techniques for detecting hybrids using Species Specific Primers (SSP’s) and High Resolution Melt (HRM). The Sanger sequencing method and ITS detected cryptic hybrids. The HRM was able to detect hybrids when comparing F1 putative hybrids to a ‘pure’ parent populations, but was not able to detect between a ‘pure’ parent population and F3 or F4 putative hybrids in the Van Stadens system. The SSP’s had the highest hybrid detection rate. Genetic methods, specifically the SSP’s were found to work well to detect hybrids in an environment where a hybrid swarm has taken place. The maps produced from this thesis will be useful for determining the distribution of hybrids in the reserve and for other similar hybrid systems. Long distance dispersal of Cape species may lead to the loss of genetic diversity or species as local and non-local hybridisation may occur, similar to the hybridisation between P. susannae and P. eximia in the Van Stadens Wild Flower Reserve. Genetic methods were shown to be superior to morphological hybrid detection (i.e. Sanger sequencing and SSP’s). The SSP approach tested here, which is dependent on the presence of fixed species specific SNP’s (Single Nucleotide Polymorphisms), could offer a rapid and effective method to explore hybridisation through space and time.
- Full Text:
- Date Issued: 2019
First report on carbon storage in warm-temperate mangroves of South Africa
- Authors: Johnson, Jaime Leigh
- Date: 2019
- Subjects: Mangrove ecology -- South Africa , Mangrove conservation Mangrove plants Mangrove management
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/31563 , vital:31581
- Description: Blue carbon is defined as the carbon sequestered and stored by coastal vegetated ecosystems including mangroves, seagrass and salt marsh. Research on blue carbon has rapidly escalated during the past decade following the 2009 United Nations Environment Program (UNEP) report, which highlighted the efficiency of these coastal habitats to act as carbon sinks. Mangrove forests in particular have been recognized for their significant role as carbon sinks and therefore have been prioritized in atmospheric carbon mitigation strategies. Despite global research efforts, there is still a notable knowledge gap on carbon storage potential of different mangrove tree species in underrepresented geographic regions. This study provides the first comprehensive assessment of blue carbon for a mangrove forest in a warm temperate estuary in South Africa. Carbon storage was quantified for Avicennia marina mangrove stands at the Nxaxo Estuary; using both elemental analysis and loss on ignition (LOI) techniques for sediment carbon pools and only elemental analysis for aboveground biomass (leaf litter, pneumatophores, wood).The environmental characteristics were the same at the five sites sampled and this was related to the estuary being well-mixed and permanently open to the Indian Ocean. Unexpectedly the carbon content was weakly correlated with that measured by elemental analysis. The cause of this is unknown and could potentially be an avenue for further investigation. The carbon content measured from elemental analysis was used in further calculations, as this method is reliable and meets global standards for comparisons. For the aboveground carbon pools, the pneumatophores (2.1 × 10-3 Mg C ha-1) stored significantly more carbon compared to the live trees (3.35 × 10-4 Mg C ha-1) and leaf litter (2.84 × 10-6 Mg C ha-1). The sediment carbon pool (228.05 Mg C ha-1) stored significantly more carbon than the aboveground carbon pools, which is in line with the findings of other studies from different regions. Sediment carbon storage was not related to mangrove tree population structure or sediment characteristics, such as moisture content and particle size. However, sediment carbon storage was highest at Site 2 which was located in a depositional area of the estuary. The total carbon storage is 1140.25 Mg C for the mangrove area of 9.5 the Nxaxo Estuary; this is considerably lower than what has been recorded for other regions. In South Africa, mangroves in warm--temperate regions occur at one of the southernmost limits globally and are restricted to occurring in sheltered estuaries due to the high--energy nature of the coastline. As a result, the growth and productivity of mangroves in these regions can be influenced by estuarine and climatic dynamics. Although total carbon storage was not variable between different sites at this estuary, this might not be the case on a regional scale as variability is expected to occur between estuaries within South Africa. Comparing the data collected from this study and others done globally, it is clear that the capacity for carbon storage by mangroves is variable. Therefore, regional scale blue carbon studies should be undertaken to provide estimates from direct quantification and allometric equation approaches. Carbon storage has been recognized as a valuable ecosystem service; therefore, going forward, accurate estimations of storage potential is essential, particularly for climate change mitigation and emission offset projects.
- Full Text:
- Date Issued: 2019
- Authors: Johnson, Jaime Leigh
- Date: 2019
- Subjects: Mangrove ecology -- South Africa , Mangrove conservation Mangrove plants Mangrove management
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/31563 , vital:31581
- Description: Blue carbon is defined as the carbon sequestered and stored by coastal vegetated ecosystems including mangroves, seagrass and salt marsh. Research on blue carbon has rapidly escalated during the past decade following the 2009 United Nations Environment Program (UNEP) report, which highlighted the efficiency of these coastal habitats to act as carbon sinks. Mangrove forests in particular have been recognized for their significant role as carbon sinks and therefore have been prioritized in atmospheric carbon mitigation strategies. Despite global research efforts, there is still a notable knowledge gap on carbon storage potential of different mangrove tree species in underrepresented geographic regions. This study provides the first comprehensive assessment of blue carbon for a mangrove forest in a warm temperate estuary in South Africa. Carbon storage was quantified for Avicennia marina mangrove stands at the Nxaxo Estuary; using both elemental analysis and loss on ignition (LOI) techniques for sediment carbon pools and only elemental analysis for aboveground biomass (leaf litter, pneumatophores, wood).The environmental characteristics were the same at the five sites sampled and this was related to the estuary being well-mixed and permanently open to the Indian Ocean. Unexpectedly the carbon content was weakly correlated with that measured by elemental analysis. The cause of this is unknown and could potentially be an avenue for further investigation. The carbon content measured from elemental analysis was used in further calculations, as this method is reliable and meets global standards for comparisons. For the aboveground carbon pools, the pneumatophores (2.1 × 10-3 Mg C ha-1) stored significantly more carbon compared to the live trees (3.35 × 10-4 Mg C ha-1) and leaf litter (2.84 × 10-6 Mg C ha-1). The sediment carbon pool (228.05 Mg C ha-1) stored significantly more carbon than the aboveground carbon pools, which is in line with the findings of other studies from different regions. Sediment carbon storage was not related to mangrove tree population structure or sediment characteristics, such as moisture content and particle size. However, sediment carbon storage was highest at Site 2 which was located in a depositional area of the estuary. The total carbon storage is 1140.25 Mg C for the mangrove area of 9.5 the Nxaxo Estuary; this is considerably lower than what has been recorded for other regions. In South Africa, mangroves in warm--temperate regions occur at one of the southernmost limits globally and are restricted to occurring in sheltered estuaries due to the high--energy nature of the coastline. As a result, the growth and productivity of mangroves in these regions can be influenced by estuarine and climatic dynamics. Although total carbon storage was not variable between different sites at this estuary, this might not be the case on a regional scale as variability is expected to occur between estuaries within South Africa. Comparing the data collected from this study and others done globally, it is clear that the capacity for carbon storage by mangroves is variable. Therefore, regional scale blue carbon studies should be undertaken to provide estimates from direct quantification and allometric equation approaches. Carbon storage has been recognized as a valuable ecosystem service; therefore, going forward, accurate estimations of storage potential is essential, particularly for climate change mitigation and emission offset projects.
- Full Text:
- Date Issued: 2019
Foreground simulations for observations of the global 21-cm signal
- Authors: Klutse, Diana
- Date: 2019
- Subjects: Cosmic background radiation , Astronomy -- Observations , Electromagnetic waves , Radiation, Background
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/76398 , vital:30557
- Description: The sky-averaged (global) spectrum of the redshifted 21-cm line promises to be a direct probe of the Dark Ages, the period before the first luminous sources formed and the Epoch of Reionization during which these sources produced enough ionizing photons to ionize the neutral intergalactic medium. However, observations of this signal are contaminated by both astrophysical foregrounds which are orders of magnitude brighter than the cosmological signal and by non-astrophysical and non-ideal instrumental effects. It is therefore crucial to understand all these data components and their impacts on the cosmological signal, for successful signal extraction. In this view, we investigated the impact that small scale spatial structures of diffuse Galactic foreground has on the foreground spectrum as observed by a global 21-cm observation. We simulated two different sets of observations using a realistic dipole beam model of two synchotron foreground templates that differ from each other in the small scale structure: the original 408 MHz all-sky map by Haslam et al. (1982) and a version where the calibration was improved to remove artifcats and point sources (Remazeilles et al., 2015). We generated simulated foreground spectra and modeled them using a polynomial expansion in frequency. We found that the different foreground templates have a modest impact on the simulated spectra, generate differences up to 2% in the root mean square of residual spectra after the log-polynomial best fit was subtracted out.
- Full Text:
- Date Issued: 2019
- Authors: Klutse, Diana
- Date: 2019
- Subjects: Cosmic background radiation , Astronomy -- Observations , Electromagnetic waves , Radiation, Background
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/76398 , vital:30557
- Description: The sky-averaged (global) spectrum of the redshifted 21-cm line promises to be a direct probe of the Dark Ages, the period before the first luminous sources formed and the Epoch of Reionization during which these sources produced enough ionizing photons to ionize the neutral intergalactic medium. However, observations of this signal are contaminated by both astrophysical foregrounds which are orders of magnitude brighter than the cosmological signal and by non-astrophysical and non-ideal instrumental effects. It is therefore crucial to understand all these data components and their impacts on the cosmological signal, for successful signal extraction. In this view, we investigated the impact that small scale spatial structures of diffuse Galactic foreground has on the foreground spectrum as observed by a global 21-cm observation. We simulated two different sets of observations using a realistic dipole beam model of two synchotron foreground templates that differ from each other in the small scale structure: the original 408 MHz all-sky map by Haslam et al. (1982) and a version where the calibration was improved to remove artifcats and point sources (Remazeilles et al., 2015). We generated simulated foreground spectra and modeled them using a polynomial expansion in frequency. We found that the different foreground templates have a modest impact on the simulated spectra, generate differences up to 2% in the root mean square of residual spectra after the log-polynomial best fit was subtracted out.
- Full Text:
- Date Issued: 2019
Gender stereotyping in traditional healing practices
- Authors: Mayaba, Nothemba
- Date: 2019
- Subjects: Gender identity -- Traditional healing -- Education Traditional medicines -- Primary health care Female traditional healers - - South Africa – Eastern
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/11260/1354 , vital:32262
- Description: People residing in rural areas use traditional medicines as their primary health care. They consult traditional healers. Traditional healing is a gender-based practice, although both men and women engaged in this practice. In some communities there are more female traditional healers than male traditional healers, including Mhlontlo Local Municipality. The main aim of the study was to investigate gender stereotyping in traditional healing in Mhlontlo Local Municipality. This study also investigated the relationship between male and female traditional healers, challenges faced by female traditional healers, transfer of traditional knowledge by female traditional healers and traditional healers’ unions. Snowball sampling was used to identify and select 79 traditional healers from Mhlontlo Local Municipality. Data was gathered using a questionnaire. The majority of traditional healers in the study area were females. Most of them have primary education, some with secondary education, and others with no formal education, and few with tertiary school. Most traditional healers were between 46-60 years. Female traditional healers constitute 70% of the study. The majority of them see between 11 and 20 patients per month. Most of the traditional healers had an experience of 21-30 years in traditional healing. The study also revealed that female traditional healers experience challenges in traditional healing. Diviners and herbalists work together, this denotes a relationship between them. Female traditional healers also have a problem in transferring their knowledge and registering under traditional healers’ unions. In conclusion, the study found out that traditional healing is gender stereotyped. There are more diviners than herbalists, where most diviners are females and herbalists are males. There is no transfer of knowledge among diviners, as being a diviner is an ancestral calling.
- Full Text:
- Date Issued: 2019
- Authors: Mayaba, Nothemba
- Date: 2019
- Subjects: Gender identity -- Traditional healing -- Education Traditional medicines -- Primary health care Female traditional healers - - South Africa – Eastern
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/11260/1354 , vital:32262
- Description: People residing in rural areas use traditional medicines as their primary health care. They consult traditional healers. Traditional healing is a gender-based practice, although both men and women engaged in this practice. In some communities there are more female traditional healers than male traditional healers, including Mhlontlo Local Municipality. The main aim of the study was to investigate gender stereotyping in traditional healing in Mhlontlo Local Municipality. This study also investigated the relationship between male and female traditional healers, challenges faced by female traditional healers, transfer of traditional knowledge by female traditional healers and traditional healers’ unions. Snowball sampling was used to identify and select 79 traditional healers from Mhlontlo Local Municipality. Data was gathered using a questionnaire. The majority of traditional healers in the study area were females. Most of them have primary education, some with secondary education, and others with no formal education, and few with tertiary school. Most traditional healers were between 46-60 years. Female traditional healers constitute 70% of the study. The majority of them see between 11 and 20 patients per month. Most of the traditional healers had an experience of 21-30 years in traditional healing. The study also revealed that female traditional healers experience challenges in traditional healing. Diviners and herbalists work together, this denotes a relationship between them. Female traditional healers also have a problem in transferring their knowledge and registering under traditional healers’ unions. In conclusion, the study found out that traditional healing is gender stereotyped. There are more diviners than herbalists, where most diviners are females and herbalists are males. There is no transfer of knowledge among diviners, as being a diviner is an ancestral calling.
- Full Text:
- Date Issued: 2019
Genetic detection of some tick-borne bacterial and protozoan pathogens in ticks collected in Raymond Mhlaba local municipality, Eastern Cape Province, South Africa
- Authors: Nqoro, Ayabulela
- Date: 2019
- Subjects: Tick-borne diseases Ticks
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/17084 , vital:40846
- Description: Ticks and tick-borne diseases are becoming a major life threatening concern to wildlife, domesticated animals and human health. Besides causing skin damage, ticks infestations have become a growing burden in food security, economic losses and transmitting multides of pathogens. Little data and knowledge is available regarding the occurrence of etiologic agents of tick-borne diseases in the Eastern Cape of South Africa, hence the study was conducted and aimed at screening for genetic material of Anaplasma, Rickettisia, Ehrlichia, Borrelia, Babesia and Theileria species in ticks collected in Raymond Mhlaba District at Eastern Cape, South Africa. Ticks were collected from domesticated animals in Raymond Mhlaba Municipality, and were morphologically identified and processed for DNA extraction. Ticks were chopped into bits and DNA was extracted from the samples with commercial DNA extraction kit. The extracted DNA samples was used to molecularly identify the tick as well as assess the presence of tickborne pathogens belonging to Rickettsia, Babesia, Borrelia, Anaplasma and Erhlichia, and Theileria spp. by PCR using specific primer pairs published in literature. Positive amplicons were sequenced in a commercial sequencing facility. The obtained chromatograms were edited with Geneious bioinformatics software and were subjected to BLASTn and phylogenetic analyses using MEGA7 version for evolutionary relationships with curated reference sequences in GenBank. Nine hundred and sixty two tick samples were collected from domestic animals. Collected tick samples belonged to three genera, which were the Amblyomma, Rhipicephalus and Haemophalis in decreasing order of their abundance. Screening of tick DNA samples by PCR did not show presence of Babesia, Borrelia, Anaplasma and Erhlichia. 13 Positive PCR products were observed for Rickettsia and Theileria spp.. The positive amplicons were purified, sequenced and analysed for speciation of Theileria and Rickettsia. The presence of Rickettsia was detected in 60/994 (6%) from the three genera of ticks. Phylogenetic analyses shows that the sequences obtained are phylogenetically related to members of Spotted fever group Rickettsiae. Genetic material of Theileria spp. was detected from 10/994 ticks with an overall infection of 1% obtained in Rhipicephalus genera. Analyses shows that the sequences obtained are phylogenetically related to T. orientalis complex. The finding from this study therefore expands the knowledge on recent emergence of Theileria and Rickettsia spp. in Raymond Mhlaba Municipality in Eastern Cape, South Africa
- Full Text:
- Date Issued: 2019
- Authors: Nqoro, Ayabulela
- Date: 2019
- Subjects: Tick-borne diseases Ticks
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/17084 , vital:40846
- Description: Ticks and tick-borne diseases are becoming a major life threatening concern to wildlife, domesticated animals and human health. Besides causing skin damage, ticks infestations have become a growing burden in food security, economic losses and transmitting multides of pathogens. Little data and knowledge is available regarding the occurrence of etiologic agents of tick-borne diseases in the Eastern Cape of South Africa, hence the study was conducted and aimed at screening for genetic material of Anaplasma, Rickettisia, Ehrlichia, Borrelia, Babesia and Theileria species in ticks collected in Raymond Mhlaba District at Eastern Cape, South Africa. Ticks were collected from domesticated animals in Raymond Mhlaba Municipality, and were morphologically identified and processed for DNA extraction. Ticks were chopped into bits and DNA was extracted from the samples with commercial DNA extraction kit. The extracted DNA samples was used to molecularly identify the tick as well as assess the presence of tickborne pathogens belonging to Rickettsia, Babesia, Borrelia, Anaplasma and Erhlichia, and Theileria spp. by PCR using specific primer pairs published in literature. Positive amplicons were sequenced in a commercial sequencing facility. The obtained chromatograms were edited with Geneious bioinformatics software and were subjected to BLASTn and phylogenetic analyses using MEGA7 version for evolutionary relationships with curated reference sequences in GenBank. Nine hundred and sixty two tick samples were collected from domestic animals. Collected tick samples belonged to three genera, which were the Amblyomma, Rhipicephalus and Haemophalis in decreasing order of their abundance. Screening of tick DNA samples by PCR did not show presence of Babesia, Borrelia, Anaplasma and Erhlichia. 13 Positive PCR products were observed for Rickettsia and Theileria spp.. The positive amplicons were purified, sequenced and analysed for speciation of Theileria and Rickettsia. The presence of Rickettsia was detected in 60/994 (6%) from the three genera of ticks. Phylogenetic analyses shows that the sequences obtained are phylogenetically related to members of Spotted fever group Rickettsiae. Genetic material of Theileria spp. was detected from 10/994 ticks with an overall infection of 1% obtained in Rhipicephalus genera. Analyses shows that the sequences obtained are phylogenetically related to T. orientalis complex. The finding from this study therefore expands the knowledge on recent emergence of Theileria and Rickettsia spp. in Raymond Mhlaba Municipality in Eastern Cape, South Africa
- Full Text:
- Date Issued: 2019
Germination potential of seeds harvested at the Worcester Veld Reserve
- Authors: Swart, Rudi
- Date: 2019
- Subjects: Ecological disturbances -- South Africa -- Karoo , Environmental degradation -- South Africa -- Karoo
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/43923 , vital:37084
- Description: The rangelands of the Succulent Karoo and Nama Karoo biomes are in various states of degradation. The injudicious stocking of overly high numbers of domestic livestock is considered to be the anthropogenic cause of this degradation. The palatable plants of these rangelands have been greatly reduced in number or extirpated from some areas. In order to return palatable plants to degraded rangelands and improve rangeland productivity, many restoration projects have been attempted that involve reseeding. The sowing of seeds into degraded rangelands, or other disturbed areas, has often provided disappointing germination results. This study will attempt to determine what some of the main aspects are that affect Karoo seed germination, both positively and negatively. The specific objectives of the study are to determine the causes of pre-sowing seed mortality, the effect of drying on seed germination, the effect of planting depth on seed germination and the effect of storage time on seed germination. The seeds of four palatable Karoo plant species harvested at the Worcester Veld Reserve were selected to be subjected to germination trails and viability testing. This study uses germination trials under controlled light and temperature conditions, as well as 2, 3, 5-triphenyl tetrazolium chloride tests to determine whether the four species of Karoo seeds harvested at the Worcester Veld Reserve for rangeland reseeding projects are viable and germinable. The practice of drying seeds after harvesting is tested to determine the effect of drying on seed germination. The germination of seed was also tested over 17 intervals of 3 weeks to determine the effect of storage time on seed germinability. Germination trials were also conducted in a nursery to determine whether there was a difference in seedling emergence between seeds planted at 10 mm depth compared to seeds planted on the surface with a partial covering of sand. The seed viability of all four species studied was found to be below 50%. Drying only significantly improved the germination of Osteospermum sinuatum seeds. The seeds of O. sinuatum and Eriocephalus africanus germinated reasonably well throughout the 17 time intervals, while the seeds of Chaetobromus involucratus and Gorteria integrifolia germinated poorly during the first six months after harvesting, after which v germination improved markedly. Seedling emergence of all four species studied was significantly higher when planted at the substrate surface, compared to seeds planted at 10 mm depth. The findings of this study show that while the viability of the seeds harvested for rangeland reseeding projects are quite low, the appropriate pre-treatment, seed age and planting depth will allow a significant increase in germination. This increase in germination should provide a greater chance of establishing palatable plants in rangeland reseeding projects.
- Full Text:
- Date Issued: 2019
- Authors: Swart, Rudi
- Date: 2019
- Subjects: Ecological disturbances -- South Africa -- Karoo , Environmental degradation -- South Africa -- Karoo
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10948/43923 , vital:37084
- Description: The rangelands of the Succulent Karoo and Nama Karoo biomes are in various states of degradation. The injudicious stocking of overly high numbers of domestic livestock is considered to be the anthropogenic cause of this degradation. The palatable plants of these rangelands have been greatly reduced in number or extirpated from some areas. In order to return palatable plants to degraded rangelands and improve rangeland productivity, many restoration projects have been attempted that involve reseeding. The sowing of seeds into degraded rangelands, or other disturbed areas, has often provided disappointing germination results. This study will attempt to determine what some of the main aspects are that affect Karoo seed germination, both positively and negatively. The specific objectives of the study are to determine the causes of pre-sowing seed mortality, the effect of drying on seed germination, the effect of planting depth on seed germination and the effect of storage time on seed germination. The seeds of four palatable Karoo plant species harvested at the Worcester Veld Reserve were selected to be subjected to germination trails and viability testing. This study uses germination trials under controlled light and temperature conditions, as well as 2, 3, 5-triphenyl tetrazolium chloride tests to determine whether the four species of Karoo seeds harvested at the Worcester Veld Reserve for rangeland reseeding projects are viable and germinable. The practice of drying seeds after harvesting is tested to determine the effect of drying on seed germination. The germination of seed was also tested over 17 intervals of 3 weeks to determine the effect of storage time on seed germinability. Germination trials were also conducted in a nursery to determine whether there was a difference in seedling emergence between seeds planted at 10 mm depth compared to seeds planted on the surface with a partial covering of sand. The seed viability of all four species studied was found to be below 50%. Drying only significantly improved the germination of Osteospermum sinuatum seeds. The seeds of O. sinuatum and Eriocephalus africanus germinated reasonably well throughout the 17 time intervals, while the seeds of Chaetobromus involucratus and Gorteria integrifolia germinated poorly during the first six months after harvesting, after which v germination improved markedly. Seedling emergence of all four species studied was significantly higher when planted at the substrate surface, compared to seeds planted at 10 mm depth. The findings of this study show that while the viability of the seeds harvested for rangeland reseeding projects are quite low, the appropriate pre-treatment, seed age and planting depth will allow a significant increase in germination. This increase in germination should provide a greater chance of establishing palatable plants in rangeland reseeding projects.
- Full Text:
- Date Issued: 2019
Grabcuts for image segmentation: a comparative study of clustering techniques
- Authors: Manzi, Nozuko Zuleika
- Date: 2019
- Subjects: Algorithms , Computer graphics
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/14494 , vital:39995
- Description: Image segmentation is the partitioning of a digital image into small segments such as pixels or sets of pixels. It is significant as it allows for the visualization of structures of interest, removing unnecessary information. In addition, image segmentation is used in many fields like, for instance healthcare for image surgery, construction, etc. as it enables structure analysis. Segmentation of images can be computationally expensive especially when a large dataset is used, thus the importance of fast and effective segmentation algorithms is realised. This method is used to locate objects and boundaries (i.e. foreground and background) in images. The aim of this study is to provide a comparison of clustering techniques that would allow the Grabcuts for image segmentation algorithm to be effective and inexpensive. The Grabcuts based method, which is an extension of the graph cut based method, has been instrumental in solving many problems in computer vision i.e. image restoration, image segmentation, object recognition, tracking and analysis. According to Ramirez,et.al [47], the Grabcuts approach is an iterative and minimal user interaction algorithm as it chooses a segmentation by iteratively revising the foreground and background pixels assignments. The method uses min-cut/ max-flow algorithm to segment digital images proposed by Boykov and Jolly [9]. The input of this approach is a digital image with a selected v region of interest (ROI). The ROI is selected using a rectangular bounding box. The pixels inside the bounding box are assigned to the foreground, while the others are assigned to the background. In this study, the Grabcuts for image segmentation algorithm designed by [48] with a Gaussian Mixture Model (GMM) based on the Kmeans and Kmedoids clustering techniques are developed and compared. In addition, the algorithms developed are allowed to run on the Central Processing Unit (CPU) under two scenarios. Scenario 1 involves allowing the Kmeans and Kmedoids clustering techniques to the Squared Euclidean distance measures to calculate the similarities and dissimilarities in pixels in an image. In scenario 2, the Kmeans and Kmedoids clustering techniques will use the City Block distance measure to calculate similarities as well as dissimilarities between pixels in a given image. The same images from the Berkeley Segmentation Dataset and Benchmark 500 were used as input to the algorithms and the number of clusters, K, was varied from 2 to 5. It was observed that the Kmeans clustering technique outperformed the Kmedoids clustering technique under the two scenarios for all the test images with K varied from 2 to 5, in terms of runtime required. In addition, the Kmeans clustering technique obtained more compact and separate clusters under scenario 1, than its counterpart. On the other hand, the Kmedoids obtained more compact and separate clusters than the Kmeans clustering technique under scenario 2. The silhouette validity index favoured the smallest number of clusters for both clustering techniques as it suggested the optimal number of clusters for the Kmeans and Kmedoids clustering techniques under the two scenarios was 2. Although the Kmeans required less computation time than vi its counterpart, the generation of foreground and background took longer for the GMM based on Kmeans than it did for the GMM based on Kmedoids clustering technique. Furthermore, the Grabcuts for image segmentation algorithm with a GMM based on the Kmedoids clustering technique was computationally less expensive than the Grabcuts for image segmentation algorithm with a GMM based on the Kmeans clustering technique. This was observed to be true under both scenario 1 and 2. The Grabcuts for image with the GMM based on the Kmeans clustering techniques obtained slightly better segmentation results when the visual quality is concerned, than its counterpart under the two scenarios considered. On the other hand, the BFscores showed that the Grabcuts for image segmentation algorithm with the GMM based on Kmedoids produces images with higher BF-scores than its counterpart when K was varied from 2 to 5 for most of the test images. In addition, most of the images obtained the majority of their best segmentation results when K=2. This was observed to be true under scenario 1 as well as scenario 2. Therefore, the Kmedoids clustering technique under scenario 2 with K=2 would be the best option for the segmentation of difficult images in BSDS500. This is due to its ability to generate GMMs and segment difficult images more efficiently (i.e. time complexity, higher BF-scores, more under segmented rather than over segmented images, inter alia.) while producing comparable visual segmentation results to those obtained by the Grabcuts for image segmentation: GMM-Kmeans.
- Full Text:
- Date Issued: 2019
- Authors: Manzi, Nozuko Zuleika
- Date: 2019
- Subjects: Algorithms , Computer graphics
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/14494 , vital:39995
- Description: Image segmentation is the partitioning of a digital image into small segments such as pixels or sets of pixels. It is significant as it allows for the visualization of structures of interest, removing unnecessary information. In addition, image segmentation is used in many fields like, for instance healthcare for image surgery, construction, etc. as it enables structure analysis. Segmentation of images can be computationally expensive especially when a large dataset is used, thus the importance of fast and effective segmentation algorithms is realised. This method is used to locate objects and boundaries (i.e. foreground and background) in images. The aim of this study is to provide a comparison of clustering techniques that would allow the Grabcuts for image segmentation algorithm to be effective and inexpensive. The Grabcuts based method, which is an extension of the graph cut based method, has been instrumental in solving many problems in computer vision i.e. image restoration, image segmentation, object recognition, tracking and analysis. According to Ramirez,et.al [47], the Grabcuts approach is an iterative and minimal user interaction algorithm as it chooses a segmentation by iteratively revising the foreground and background pixels assignments. The method uses min-cut/ max-flow algorithm to segment digital images proposed by Boykov and Jolly [9]. The input of this approach is a digital image with a selected v region of interest (ROI). The ROI is selected using a rectangular bounding box. The pixels inside the bounding box are assigned to the foreground, while the others are assigned to the background. In this study, the Grabcuts for image segmentation algorithm designed by [48] with a Gaussian Mixture Model (GMM) based on the Kmeans and Kmedoids clustering techniques are developed and compared. In addition, the algorithms developed are allowed to run on the Central Processing Unit (CPU) under two scenarios. Scenario 1 involves allowing the Kmeans and Kmedoids clustering techniques to the Squared Euclidean distance measures to calculate the similarities and dissimilarities in pixels in an image. In scenario 2, the Kmeans and Kmedoids clustering techniques will use the City Block distance measure to calculate similarities as well as dissimilarities between pixels in a given image. The same images from the Berkeley Segmentation Dataset and Benchmark 500 were used as input to the algorithms and the number of clusters, K, was varied from 2 to 5. It was observed that the Kmeans clustering technique outperformed the Kmedoids clustering technique under the two scenarios for all the test images with K varied from 2 to 5, in terms of runtime required. In addition, the Kmeans clustering technique obtained more compact and separate clusters under scenario 1, than its counterpart. On the other hand, the Kmedoids obtained more compact and separate clusters than the Kmeans clustering technique under scenario 2. The silhouette validity index favoured the smallest number of clusters for both clustering techniques as it suggested the optimal number of clusters for the Kmeans and Kmedoids clustering techniques under the two scenarios was 2. Although the Kmeans required less computation time than vi its counterpart, the generation of foreground and background took longer for the GMM based on Kmeans than it did for the GMM based on Kmedoids clustering technique. Furthermore, the Grabcuts for image segmentation algorithm with a GMM based on the Kmedoids clustering technique was computationally less expensive than the Grabcuts for image segmentation algorithm with a GMM based on the Kmeans clustering technique. This was observed to be true under both scenario 1 and 2. The Grabcuts for image with the GMM based on the Kmeans clustering techniques obtained slightly better segmentation results when the visual quality is concerned, than its counterpart under the two scenarios considered. On the other hand, the BFscores showed that the Grabcuts for image segmentation algorithm with the GMM based on Kmedoids produces images with higher BF-scores than its counterpart when K was varied from 2 to 5 for most of the test images. In addition, most of the images obtained the majority of their best segmentation results when K=2. This was observed to be true under scenario 1 as well as scenario 2. Therefore, the Kmedoids clustering technique under scenario 2 with K=2 would be the best option for the segmentation of difficult images in BSDS500. This is due to its ability to generate GMMs and segment difficult images more efficiently (i.e. time complexity, higher BF-scores, more under segmented rather than over segmented images, inter alia.) while producing comparable visual segmentation results to those obtained by the Grabcuts for image segmentation: GMM-Kmeans.
- Full Text:
- Date Issued: 2019