Livestock water productivity: towards improving rural livelihoods from livestock in semi-arid rangelands
- Authors: Gusha, Bukho
- Date: 2019
- Subjects: Livestock -- South Africa -- Eastern Cape , Livestock -- Effect of drought on -- South Africa -- Eastern Cape , Animals -- Food -- South Africa -- Eastern Cape , Livestock -- Effect of water quality on -- South Africa -- Eastern Cape , Livestock -- Water requirements -- South Africa -- Eastern Cape , Livestock productivity -- South Africa -- Eastern Cape , Stochastic analysis , Communal rangelands -- South Africa -- Eastern Cape , Land degradation -- South Africa -- Eastern Cape , Animal owners -- South Africa -- Eastern Cape , Livestock improvement -- South Africa -- Eastern Cape
- Language: English
- Type: text , Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10962/115171 , vital:34084
- Description: Communal rangelands in South Africa mainly occur in the former homelands. The former homelands constitute 13% of the land surface area and support a quarter of the country's human population with a wide range of goods and services, among them, grazing for livestock, mostly reared on communal rangelands. These rangelands are degraded and cannot sustain maximum livestock production because of poor species composition and low standing biomass, however research has been conducted on livestock production at household level (where all livestock goods and services are valued). This provides an opportunity to conduct a study describing livestock water productivity in the north of the Eastern Cape, where livestock production is a primary source of livelihood for rural communities from which many households generate cash but where different practices and factors undermine high livestock production. Many studies have focused on understanding the water productivity of a natural rangeland system for commercially oriented crop-livestock systems, but the aim of this study is to contribute towards improving rural livelihoods from livestock in the sub-humid rangelands of the north Eastern Cape. Here, unimproved native grasslands are the major source of feed for livestock and people do not have herders to take livestock to the most productive parts of the rangelands. Households were surveyed using a questionnaire on livestock household contribution, socioeconomic characteristics of the household, livestock holdings and livestock production strategies. Rangeland productivity was measured in the field. Experimental animals for livestock grazing distribution were identified and fitted with Global Positioning Systems (GPS) collars to identify the seasonal grazing areas. These activities shed light on the biophysical attributes of the ecosystem and livestock production in a communal rangeland system. Because continuous grazing in the rangelands of the north Eastern Cape reduces the standing biomass, there is no obvious aboveground biomass to provide a visual perspective of production nor is it possible to determine production without excluding the livestock. Thus, four parallel lines of evidence were employed to measure rangeland productivity: line intercept, grazing exclosures, net photosynthesis from earth observation and disc pasture meter. Earth observation products were used to derive the amount of water used by the landscape to produce this forage (i.e. evapotranspiration or ET) and these measurements of net primary production and landscape water use were used in preparing a value of livestock water productivity (LWP) for this farming system. There has been the perception that residents of the study area lack knowledge of technical efficiencies in the large stock sector at household level. The study used stochastic frontier analysis to assess livestock production and followed with a household survey to collect information on socio-economic characteristics and information on livestock practices. The data from the household survey were used to estimate the technical efficiency of households using a stochastic frontier analysis. Productivity and inefficiency variables that increase livestock production or increase technical difficulties were identified. The focus on livestock has mostly been on the direct value of livestock to owners with a poor understanding of their value to non-livestock owners, where cultural activities, such as livestock slaughtering, were documented as the only source of protein for non-livestock owners. However, the value that is available to non-livestock owners has not been quantified. This study assessed livestockbased livelihoods of communal people to improve their livelihoods through a household survey looking at the contribution of livestock to both livestock and non-livestock owners. Earlier work on LWP has focused on systems where animals were on ‘fed, cut and carry’ and irrigated systems. However, there is a need to describe LWP in a natural grazing system and this study set out to achieve this for these communal rangelands through a household survey that determined the value of livestock goods and services given the amount of water used (ET). Lastly, livestock grazing distribution across the landscape was assessed, using GPS collars that recorded livestock behaviour every five minutes during the daylight. This approach was necessary because livestock grazing patterns in these communal rangelands is poorly controlled by people, and animals are largely free-ranging, grazing selectively, based on their own preferences, which leads to localised overgrazing. This part of the study was achieved through experimental livestock collaring and weighing (both sheep and goats) for the wet and dry seasons. The collared livestock were weighed on the day of putting on collars and the day of removing the collars. The results on livestock grazing distribution were analysed using the R package, T-LoCoH. The major finding of this study was that communal rangelands of the north Eastern Cape can improve rural livelihoods from livestock if proper interventions for both livestock and rangeland production and productivity can be implemented. One of these interventions is fencing as it was found that exclosures that were fenced during the study yielded high aboveground productivity comparable to that achieved in commercial rangelands, yielding 220 g DM m-2 yr-1. Surveys using the calibrated disc pasture meter showed the need for proper rotation and resting of the rangeland. Net photosynthesis of 880.7 g C m-2 yr-1 for unimproved grassland in good condition was comparable to commercial rangelands in the region. Using the line intercept, vegetation cover was found to be a good predictor of aboveground standing biomass; thus a positive relationship was revealed. Lastly, annual ET of 270 mm yr-1 was calculated using the Penman Monteith Palmer (PMP) equation, while 379 mm yr-1 was extracted from the MOD16 product, suggesting that PMP ET may not be accurate in these grassland systems due to the slow response of MODIS Leaf Area Index (LAI). The average household technical efficiency (TE) score was found to be 0.79 on the study sites, indicating the potential for households to improve outputs from livestock. A range of household categories were identified, based on gender and an index of wealth, and households with lower and higher TE were identified. This analysis revealed that productivity variables such as holding higher livestock numbers and providing additional feed achieved high livestock outputs, suggesting high livestock productivity. However, in terms of inefficiency variables, gender (female-headed households), dwelling type (an index of homestead wealth), kraaling livestock at night and herding livestock during the day were found to improve technical efficiency. It was revealed in this study that households keep livestock to derive different goods and services including offtake, manure, milk, wool and services such as traction. The non-livestock owning households were reported to also benefit from the abovementioned goods and services in the study site and that the value of their contribution could be quantified, thus contributing significantly to rural livelihoods. The study showed that LWP was comparable with other studies such as those conducted in Ethiopia. This study compared its results with the studies conducted outside South Africa as there were limited comparable South African studies available; however, this does not necessarily mean we can use the same model as the value of livestock outputs varies based on the preferred outputs. This study developed an LWP model for the natural rangeland system. The LWP values were measured in ZAR and later converted in USD and were divided into three different categories based on the wealth index, such as better-off, middle wealth and poor households. Lastly, this study showed that livestock (both cattle and sheep) spend a high proportion of their grazing day, during both the wet and dry seasons, in a small physical area, immediately around the homesteads. These are areas where the active green growth occurs throughout the year, suggesting the need for livestock herders to move livestock around the landscape for more effective landscape use. Herding has the potential to improve landscape use and conserve grazing resource and the ability of a household to attain best outputs from livestock. Positive daily weight gains were reported in collared livestock during the wet season. However, both sheep and cattle lost weight during the dry season. This study recommends interventions such as labour for herding, and other animal husbandry-related activities including milking, handling, and vaccinating animals. Market opportunities for communal rangeland livestock should be facilitated by informing livestock owners about livestock market specifications to improve their livelihoods. Lastly, proper grazing management planning, such as fencing, which enables rotational grazing, and herding which moves animals to the most productive parts of the rangeland, should be implemented so that rangelands can be rested for plant growth, vigour, and improved aboveground net primary productivity. Based on the recommendations made in this study, a research development approach is necessary which prioritises female empowerment in agriculture and poor farmers as female-headed households were reported by this study to be more technically efficient.
- Full Text:
- Date Issued: 2019
Woody encroachment and plant-root soil interactions in a semi-arid savanna
- Authors: Gusha, Bukho
- Date: 2015
- Subjects: Pastures -- Management Weeds Invasive plants
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/12139 , vital:39174
- Description: Savannas occupy 1/8 of the global land surface, support a large proportion of the world’s human population and the majority of its rangeland and livestock. Woody encroachment has been reported as the major challenge in these landscapes. This study describes the differences between three contrasting tree density classes in a semi- arid savanna with the view to developing an improved understanding of woody encroachment which is prevalent in this region. The study attempted to determine if there was a relationship between lateral root distributions at varying soil depths with increasing levels of woody encroachment, and to compare species composition and soil water profiles in these rangelands. Three homogeneous vegetation units, namely: sparsely encroached (HVU1), grassland (HVU2) and Albany thicket (HVU3) were identified for the study sites. A Trench method was used determine root biomass and a step point method was used to determine herbaceous species composition in all the HVUs. The results showed that more Decreaser species (especially Themeda triandra) were recorded in a sparsely encroached site (HVU1) and grassland site (HVU2), while Cynodon dactylon was mostly recorded in the thicket site (HVU3). Acacia karroo was mostly recorded in HVU1 while in HVU2 other woody species such as Coddia rudis and Grewia occidentalis were also recoded. HVU3 had the poorest basal cover (point to tuft distance) (22.60 cm) while HVU1 (9.93 cm) and HVU2 (7.73 cm) had moderate basal cover. The herbaceous standing biomass was significantly different across the HVUs. HVU3 was higher (1206.15 kg ha-1) than HVU1 (942.43 kg ha-1) and HVU2 (677.10 kg ha-1). The soil moisture content was significantly different between the HVUs, but was not significantly different between the depths (p< 0.05). High soil moisture content was recorded in HVU3 compared to other HVUs. The results of Analysis of Variance (ANOVA) showed that soil depth and the type of homogenous vegetation unit (HVU) had significant effects (p<0.01) on root biomass. The pair wise t-test showed that there were no significant difference in root biomass between sparsely encroached (HVU1) and grassland (HVU2) sites (p>0.05), but there were significant differences in root biomass between grassland (HVU2) and thicket (HVU3) site (p<0.05). The mean for total root biomass found in the study was 2.66 kg m-2. In all the trenches most of the root biomass was found in depth 1 (0-30 cm) which was 2.43 kg m-2followed by 1.32 kg m-2 in depth 2 (30-60 cm) and 0.49 kg m-2 in depth 3 (60-90 cm). According to the results on species composition, herbaceous biomass, basal cover, soil moisture content and the root biomass, Kwezana communal rangeland has a potential of running a sustainable livestock production enterprise if proper management practices can be implemented. To improve the rangelands of Kwezana communal rangelands, management such as proper resting, burning, proper stocking rates and physically clearing of bushes should be considered.
- Full Text:
- Date Issued: 2015
Woody encroachment and plant-root soil interactions in a semi-arid savanna
- Authors: Gusha, Bukho
- Date: 2015
- Subjects: Pastures -- Management Weeds Invasive plants
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10353/12053 , vital:39131
- Description: Savannas occupy 1/8 of the global land surface, support a large proportion of the world’s human population and the majority of its rangeland and livestock. Woody encroachment has been reported as the major challenge in these landscapes. This study describes the differences between three contrasting tree density classes in a semi- arid savanna with the view to developing an improved understanding of woody encroachment which is prevalent in this region. The study attempted to determine if there was a relationship between lateral root distributions at varying soil depths with increasing levels of woody encroachment, and to compare species composition and soil water profiles in these rangelands. Three homogeneous vegetation units, namely: sparsely encroached (HVU1), grassland (HVU2) and Albany thicket (HVU3) were identified for the study sites. A Trench method was used determine root biomass and a step point method was used to determine herbaceousspecies composition in all the HVUs. The results showed that more Decreaser species (especially Themeda triandra) were recorded in a sparsely encroached site (HVU1) and grassland site (HVU2), while Cynodon dactylon was mostly recorded in the thicket site (HVU3). Acacia karroo was mostly recorded in HVU1 while in HVU2 other woody species such as Coddia rudis and Grewia occidentalis were also recoded. HVU3 had the poorest basal cover (point to tuft distance) (22.60 cm) while HVU1 (9.93 cm) and HVU2 (7.73 cm) had moderate basal cover. The herbaceous standing biomass was significantly different across the HVUs. HVU3 was higher (1206.15 kg ha-1) than HVU1 (942.43 kg ha-1) and HVU2 (677.10 kg ha-1). The soil moisture content was significantly different between the HVUs, but was not significantly different between the depths (p< 0.05). High soil moisture content was recorded in HVU3 compared to other HVUs. The results of Analysis of Variance (ANOVA) showed that soil depth and the type of homogenous vegetation unit (HVU) had significant effects (p<0.01) on root biomass. The pair wise t-test showed that there were no significant difference in root biomass between sparsely encroached (HVU1) and grassland (HVU2) sites (p>0.05), but there were significant differences in root biomass between grassland (HVU2) and thicket (HVU3) site (p<0.05). The mean for total root biomass found in the study was 2.66 kg m-2. In all the trenches most of the root biomass was found in depth 1 (0-30 cm) which was 2.43 kg m-2followed by 1.32 kg m-2 in depth 2 (30-60 cm) and 0.49 kg m-2 in depth 3 (60-90 cm). According to the results on species composition, herbaceous biomass, basal cover, soil moisture content and the root biomass, Kwezana communal rangeland has a potential of running a sustainable livestock production enterprise if proper management practices can be implemented. To improve the rangelands of Kwezana communal rangelands management such as proper resting, burning, proper stocking rates and physically clearing of bushes should be considered.
- Full Text:
- Date Issued: 2015