AGRO-ECOLOGICAL RANGELAND CONDITION ASSESSMENT OF EXTENSIVE LAND-REFORM PASTORAL FARMING SITES IN THE BLOEMFONTEIN MAGISTERIAL AREA POST TWO WET SEASONS PRECEDED BY DROUGHT By THABISO EMMANUEL MOKHESENGOANE Thesis submitted in fulfilment of the requirements for the degree PHILOSOPHIAE DOCTOR (SUSTAINABLE AGRICULTURE) in the FACULTY OF NATURAL AND AGRICULTURAL SCIENCES Department of Sustainable Food Systems and Development University of the Free State Supervisor: Prof J.A Van Niekerk Co-supervisor: Dr H.C Van. der Westhuizen November 2023 DECLARATION I, Thabiso Emmanuel Mokhesengoane, declare that the thesis hereby submitted by me for the PHILOSOPHIAE DOCTOR (SUSTAINABLE AGRICULTURE) degree at the University of the Free State is my own independent work and has not previously been submitted by me at another university/faculty. ____________________ __________________ Thabiso Emmanuel Mokhesengoane Date i DEDICATION I dedicate this work to my late grandparents Mr. Sello John and Mrs. Tselane Selina Mokhesengoane who were passionate about education. ii ACKNOWLEDGEMENTS To my supervisors, I reiterate the words of Sir Isaac Newton “Indeed if I have seen further in life, it is because I stood on the shoulders of the giants”. To my mother Palesa Pretty “Ausi Ma-Thabiso” thank you for not giving up on me even though I gave you so many reasons to do so. To my family, my wife Lesimole “Mpai” and the boys Kananelo and Thabiso Junior thank you for your understanding, unwavering support and for always being my pillars of strength during tough times. Last but not least, a big thank you to Bloemfontein land-reform farmers for their participation in this research project. To Almighty God be the glory. iii TABLE OF CONTENTS DECLARATION ................................................................................................................... LIST OF TABLES ............................................................................................................. vii LIST OF FIGURES .......................................................................................................... viii LIST OF ABBREVIATIONS AND ACRONYMS ................................................................. x ABSTRACT ...................................................................................................................... xii CHAPTER 1 INTRODUCTION AND BACKGROUND TO THE STUDY ................................ 1 1.1 INTRODUCTION .................................................................................................... 1 1.2 BACKGROUND OF THE STUDY ........................................................................... 7 1.3 RESEARCH MOTIVATION ..................................................................................... 8 1.4 RESEARCH AIM .................................................................................................. 10 1.5 RESEARCH OBJECTIVES ................................................................................... 10 1.6 RESEARCH QUESTIONS .................................................................................... 11 1.7 HYPOTHESIS ....................................................................................................... 11 CHAPTER 2 LITERATURE REVIEW .................................................................................. 12 2.1 INTRODUCTION .................................................................................................. 12 2.2 RANGELANDS MANAGEMENT ........................................................................... 12 2.3 GRASSLANDS VEGETATION.............................................................................. 14 2.4 LIVESTOCK REPRODUCTION PERFORMANCE ................................................ 15 2.5 LAND DEGRADATION ......................................................................................... 16 2.6 CLIMATE VARIATION .......................................................................................... 17 2.7 SOIL PROPERTIES .............................................................................................. 18 2.8 POLICY ................................................................................................................ 20 2.9 SOUTH AFRICAN LAND-REFORM HISTORY ..................................................... 20 2.10 CONCLUSION .................................................................................................. 23 CHAPTER 3 MATERIALS AND METHODS ........................................................................ 25 3.1 STUDY AREA/LOCATION .................................................................................... 25 3.2 RESEARCH METHODOLOGY ............................................................................. 26 3.3 RESEARCH INSTRUMENTS ............................................................................... 27 3.4 SAMPLING PROCEDURE .................................................................................... 28 3.5 SAMPLE ............................................................................................................... 28 3.6 DATA COLLECTION ............................................................................................ 28 3.7 DATA ANALYSIS .................................................................................................. 29 3.8 ETHICAL CONSIDERATIONS .............................................................................. 30 CHAPTER 4 RESULTS AND DISCUSSIONS ..................................................................... 31 iv 4.1 PARTICIPANTS OVERVIEW ................................................................................ 31 4.2 SPECIFIC INDIVIDUAL PARTICIPANT NARRATIVE ........................................... 31 4.2.1 Participant 1 ................................................................................................... 31 4.2.1.1 Production level ...................................................................................... 32 4.2.1.2 Financial implications .............................................................................. 32 4.2.2 Participant 2 ................................................................................................... 32 4.2.2.1 Production level ...................................................................................... 33 4.2.2.2 Financial implications .............................................................................. 33 4.2.3 Participant 3 ................................................................................................... 33 4.2.3.1 Production level ...................................................................................... 34 4.2.3.2 Financial implications .............................................................................. 34 4.2.4 Participant 4 ................................................................................................... 34 4.2.4.1 Production level ...................................................................................... 35 4.2.4.2 Financial implications .............................................................................. 35 4.2.5 Participant 5 ................................................................................................... 35 4.2.5.1 Production level ...................................................................................... 36 4.2.5.2 Financial implications .............................................................................. 36 4.2.6 Participant 6 ................................................................................................... 37 4.2.6.1 Production level ...................................................................................... 37 4.2.6.2 Financial implications .............................................................................. 37 4.2.7 Participant 7 ................................................................................................... 38 4.2.7.1 Production level ...................................................................................... 38 4.2.7.2 Financial implications .............................................................................. 38 4.2.8 Participant 8 ................................................................................................... 39 4.2.8.1 Production level ...................................................................................... 39 4.2.8.2 Financial implications .............................................................................. 40 4.2.9 Participant 9 ................................................................................................... 40 4.2.9.1 Production level ...................................................................................... 40 4.2.9.2 Financial implications .............................................................................. 41 4.3 COLLECTIVE SYNOPSIS OF PARTICIPANTS’ PRODUCTION LEVELS ANALYSIS ............................................................................................................................. 41 4.4 RESULTS AND DISCUSSIONS ........................................................................... 42 4.4.1 Variations in land size, utilization and minimum and maximum stocking rates for sampled extensive land-reform livestock farmers ........................................................ 42 4.4.2 Variations in rangeland condition scores, including the average, minimum and maximum scores for sampled extensive land-reform livestock farmers ....................... 43 v 4.4.3 Variations in soil pH (KCI) distributions represented by the mean, std. deviations, minimum and maximum compared with rangeland condition scores among the camps .. ...................................................................................................................... 45 4.4.4 Variations in chemical soil properties distribution represented by the mean, std. deviation, maximum and minimum concentration ........................................................ 48 4.4.4.1 Phosphorus ............................................................................................ 48 4.4.4.2 Potassium ............................................................................................... 49 4.4.4.3 Calcium................................................................................................... 50 4.4.4.4 Magnesium ............................................................................................. 51 4.4.4.5 Sodium ................................................................................................... 52 4.4.4.6 Soil carbon .............................................................................................. 53 4.4.5 Variations in EA, CEC, and EC represented by the mean std. deviation, maximum and minimum concentration ........................................................................ 55 4.4.6 Variations in three soil fractions, sand, clay, and silt distribution, are represented by the mean, std. deviation, maximum and minimum concentration ............................ 58 4.4.7 Comparison between soil chemical properties, soil texture and rangeland condition scores represented by the mean, std. deviation, maximum and minimum .... 62 4.4.8 Comparison between rangeland condition scores and production planning as management input represented by the mean and std. deviation .................................. 63 4.4.9 Comparison between rangeland condition scores between camps on a particular farm and between the farms represented by the mean, std. deviation, maximum and minimum ..................................................................................................................... 64 4.5 ESSENTIAL PARAMETERS DISCUSSION .......................................................... 65 4.5.1 Rangeland condition scores and stocking rates ............................................. 65 4.5.2 Soil pH (KCl) .................................................................................................. 66 4.5.3 Soil chemical properties ................................................................................. 66 4.5.4 Rangeland management ................................................................................ 67 4.5.5 Livestock management and Reproduction performance ................................ 67 CHAPTER 5 CONCLUSION AND RECOMMENDATIONS ................................................. 69 5.1 INTRODUCTION .................................................................................................. 69 5.2 CONCLUSION ...................................................................................................... 69 5.3 RECOMMENDATIONS ......................................................................................... 71 5.4 FUTURE RESEARCH RECOMMENDATIONS ..................................................... 71 5.5 STUDY LIMITATIONS .......................................................................................... 72 5.6 STUDY CONTRIBUTIONS ................................................................................. 722 REFERENCES................................................................................................................ 74 CONFERENCE CONTRIBUTIONS AND SCIENTIFIC PUBLICATIONS ......................... 89 vi APPENDICES ............................................................................................................... 124 APPENDIX A: QUESTIONNAIRE ............................................................................. 124 APPENDIX B: PHOTOS ............................................................................................ 136 APPENDIX C: APPROVED RESEARCH ETHICAL CLEARANCES.......................... 138 APPENDIX D: SOIL ANALYSES RESULTS AND RANGELAND CONDITION SCORES .................................................................................................................................. 140 APPENDIX E: BEEF WEEKLY MARKET .................................................................. 141 vii LIST OF TABLES Table 4.1: Descriptive statistical analyses regarding the total farm size, grazing area and arable land .......................................................................................................................... 42 Table 4.2: Descriptive statistics of rangeland condition scores ............................................ 43 Table 4.3: Descriptive statistical analyses of soil pH (KCl) .................................................. 45 Table 4.4: Soil chemical properties ..................................................................................... 48 Table 4.5: Topsoil carbon content (%) for different rangelands condition classes ................ 54 Table 4.6: EA, CEC and EC ................................................................................................ 55 Table 4.7: Three fractions of soil texture determination ....................................................... 58 Table 4.8: Correlation between rangeland condition and soil chemical properties ............... 62 Table 4.9: Average rangeland condition scores and production plan ................................... 63 Table 4.10: Comparison of rangeland condition between camps per farm .......................... 64 viii LIST OF FIGURES Figure 1.1: Bloemfontein rainfall differences report for hundred years from 1 October to 28 February from South African Weather Services (SAWS) data ............................................... 8 Figure 2.1: South African grasslands biome map (LandCover 2013/14 Data Set, 2015) ..... 14 Figure 3.1: Bloemfontein magisterial area map highlighting the position of studied farms ... 25 Figure 4.1: Histogram of rangeland condition scores on land-reform farms ......................... 44 Figure 4.2: Normal Q-Q plot of rangeland condition scores on different land reform farms illustrating normal distribution .............................................................................................. 45 Figure 4.3: Histogram of soil pH (KCl) ................................................................................. 46 Figure 4.4: Q-Q plot confirming slightly uneven soil pH (KCl) level distribution amongst camps ........................................................................................................................................... 47 Figure 4.5: The Histogram of soil Phosphorus on different land-reform farms ..................... 49 Figure 4.6: Q-Q Plot of Soil Phosphorus on different land reform farms confirming slightly abnormal distribution ........................................................................................................... 49 Figure 4.7: Histogram of soil potassium on different land reform farms ............................... 50 Figure 4.8: Q-Q plot of soil potassium on different land-reform farms confirming normal distribution .......................................................................................................................... 50 Figure 4.9: Histogram of soil calcium on different land reform farms ................................... 51 Figure 4.10: Q-Q plot of soil calcium on different land-reform farms confirming the fairly abnormal distribution ........................................................................................................... 51 Figure 4.11: Histogram of soil magnesium on different land-reform farms ........................... 52 Figure 4.12: Q-Q plot of soil magnesium on different land-reform farms confirming relatively normal distribution ............................................................................................................... 52 Figure 4.13: 19 Histogram of sodium on different land-form farms ...................................... 53 Figure 4.14: Q-Q plot of sodium distribution on different land-reform farms illustrating a slightly abnormal distribution ........................................................................................................... 53 Figure 4.15: Histogram of soil carbon illustrating normal distribution ................................... 54 Figure 4.16: Histogram of EA on different land-reform farms illustrating an abnormal distribution .......................................................................................................................... 56 Figure 4.17: Q-Q plot confirming the abnormal distribution of EA ........................................ 56 Figure 4.18: Histogram of CEC on different land-reform farms illustrating normal distribution ........................................................................................................................................... 57 Figure 4.19: Q-Q plot of CEC on different land reform farms confirms relatively normal distribution .......................................................................................................................... 57 ix Figure 4.20: Histogram of EC on different land-reform farms illustrating an abnormal distribution .......................................................................................................................... 57 Figure 4.21: Q-Q plot of EC on different land reform farms confirming the abnormal distribution ........................................................................................................................................... 58 Figure 4.22: Histogram of soil density amongst sampled farms illustrating a normal distribution ........................................................................................................................................... 59 Figure 4.23: Normal Q-Q plot of soil density amongst sampled farms, confirming normal distribution .......................................................................................................................... 59 Figure 4.24: Histogram of sandy soil fraction amongst sampled farms ................................ 60 Figure 4.25: Q-Q plot of sandy soil fraction amongst sampled farms, illustrating the normal distribution .......................................................................................................................... 60 Figure 4.26: Histogram of clayey soil fraction among sampled farms .................................. 60 Figure 4.27: Q-Q plot of clayey soil fraction amongst sampled farms, illustrating the normal distribution .......................................................................................................................... 61 Figure 4.28: 34 Histogram of silt soil fraction amongst sampled farms ................................ 61 Figure 4.29: Q-Q plot of silt soil fraction amongst sampled farms, illustrating the normal distribution .......................................................................................................................... 61 Figure 4.30: Rangeland condition comparison between sampled land-reform farms ........... 65 x LIST OF ABBREVIATIONS AND ACRONYMS ANC African National Congress. ANOVA Analysis of Variance C Carbon Ca Calcium CEC Cation Exchange Capacity CH4 Methane CO2 Carbon dioxide DAFF Department of Agriculture, Forestry & Fisheries DLA Department of Land Affairs DREA Department of Rural Economy and Agriculture EA Exchangeable acidity EC Electrical Conductivity EIV Ecological Index Value FAO Food and Agriculture Organization H2O Water Ha Hectare K Potassium KCl Potassium Chloride LRAD Land Redistribution for Agricultural Development. LSUs Large Stock Units Mg Magnesium N2O Nitrous oxide Na Sodium P Phosphorus xi PLAS Proactive Land Acquisition Strategy. SADC Southern African Development Community SAR Sodium Adsorption Ratio SDGs Sustainable Development Goals SLAG Settlement /Land Acquisition Grant SOC Soil Organic Carbon SOM Soil Organic Matter T. Triandra Themeda triandra. UN United Nations UNCCD United Nations Convention to Combat Desertification WSWB Willing seller- Willing buyer. Zn Zinc xii ABSTRACT The study aimed to provide holistic guidance for future sustainable livestock production in the semi-arid rangelands of the Bloemfontein magisterial area for extensive land reform livestock farmers. It aimed to compare rangeland condition per land reform farm and between land reform farms, and to establish correlations between rangeland condition, soil chemical properties and managerial inputs on sampled land reform farms. The study also aimed to use land reform livestock farmers’ decisions on livestock stocking rates to establish their production efficiency and to establish correlations between rangeland condition and soil carbon. Nine extensive land-reform livestock farms out of a total of twenty-nine land-reform farms which participated in the stocking rate study during the 2018/2019 drought were sampled using a non-probability snowball sampling procedure. A multi-methodological approach, namely, qualitative and quantitative methodologies, was employed to investigate the research topic. Structured questionnaires were used to determine profiles of rangeland managers, farm management information, and sound sustainable livestock management information, including current stocking rates on all nine sampled farms. The rangeland degradation gradient method for the semi-arid rangelands of the central Free State province region was used to determine rangelands condition scores together with the wheel point apparatus tool on two camps of each sampled farm. Soil samples were taken from each camp at the beginning of the rangeland condition assessment transect line, in the middle and end, at 20 cm depth, using the soil auger tool. The descriptive statistics, average scores, and standard deviations were calculated using analysis of variance (ANOVA), Pearson correlation, and t-tests to compare the average values. Statistical Package for Social Science (SPSS) version 28 was used to analyse quantitative data. Metabolic body weights were used to calculate large stock units. The study recorded a significant relationship (p< 0.05) between soil carbon (C) and rangeland condition, with three times more C on a good rangeland condition score of >60% when compared to a poor rangeland condition score of <20%. A strong relationship (p< 0.05) was also found between a production plan and higher average rangeland condition scores. However, no significant relationship (p> 0.05) was found between other soil chemical properties and rangeland condition, except for the positive trends recorded on soil phosphorus xiii (P) and soil pH levels and a significant relationship between soil carbon (C) and rangeland condition. On the other hand, no relationship (p> 0.05) was found between rangeland condition scores and other variables. The highest average rangeland condition score was 77%, and the minimum average rangeland condition score recorded amongst farms was 21%. Thus, the study concludes that the implementation of science-based knowledge on both stocking rates and rangeland management for extensive land-reform livestock farmers in the Bloemfontein magisterial area is paramount and valuable for climate change impacts mitigation and sustainable livestock productivity. Furthermore, the study recommends investigating carbon market participation as an incentive for farmers to implement sustainable ecological practices and the practicality of incorporating variable stocking rates into extensive livestock management systems. Keywords: Rangelands assessment, Land-reform, Extensive pastoral farms. 1 CHAPTER 1 INTRODUCTION AND BACKGROUND TO THE STUDY 1.1 INTRODUCTION Rangelands are generally defined as parts of the world where wildlife and domestic livestock graze on the natural vegetation (Squires, 2010; Craggs, 2017). However, according to Shackleton et al. (2001) and Dovie et al. (2002), rangelands also play an important role as the source of other natural resources, which include fuel wood, thatching grass, wild fruits, edible herbs and medicinal plants and all these are harvested for both household use and sale. Since it is clear that these rangelands have constantly been subjected to one form of management system or another, the importance of sustainable rangeland management practices for sustainable pastoral farming cannot be over-emphasized. “Pastoralism is defined as a production system in which 50% or more of the household gross income is derived from livestock or livestock-related activities” (Niamir-Fuller, 1998). Thus, pastoralism is a science of raising livestock, particularly small ruminants, cattle and camels as a primary means of survival. Extensive pastoral/livestock farming systems are standard in Africa's vast open arid and semi-arid rangelands. A distinctive feature of these ecosystems is the variable annual average rainfall precipitation rate, leading to the differences in rangeland characteristics due to differences in water resource distributions. The latter led to the development of different rangeland management systems by pastoralists, which are custom-made for each geographical region, as the pastoralists adapt to trends such as new economic opportunities and utilization of the modern means of livestock farming (DREA, 2010). The following facts have greatly influenced the socioeconomic importance of extensive pastoral farming in Africa: • Pastoral areas occupy about 43% of Africa’s land mass, albeit with significant variations between countries (FAO, 2018). • Livestock or livestock-related economic activities contribute at least 5 to 100 per cent of the total GDP and cover 15 to 40 per cent of the added value in agriculture within the dry lands of Sub-Saharan Africa (De Haan, 2016). 2 • In general, pastoral areas are less suitable for crop husbandry, and livestock production remains the most viable opportunity to harness scarce biomass resources (Coppock et al., 2017). • Over and above, there is clear evidence of rangeland degradation in recent years, resulting from human activities such as fuel wood collection and crop farming in the marginal areas (UNEP, 2000), including extensive livestock farming activities. Nonetheless, according to Child and Frasler (1992) and Mitchell (2000), these grassland ecosystem areas cover 40-50% of the world’s surface area, and Egoh et al. (2011) further confirmed that the grassland is home to approximately one billion people around the globe. However, the United Nations Convention to Combat Desertification identified rangeland degradation in arid and semi-arid grassland regions as a global phenomenon (UNCCD, 1995). The White Paper on Agricultural Policy (1984) pointed out that the South African natural rangelands are deteriorating tremendously (Du Toit et al., 1991). Thus, the deteriorating state of the natural rangelands can be attributed to human and non-human parameters. Essential human parameters, amongst others, include imprudent rangelands management practices such as higher or lower animal stocking rates; both extremes are detrimental to rangelands sustainability (Mokhesengoane et al., 2021), as they lead to over-utilization and underutilization of grazeable materials, and thus impacting rangelands species composition negatively. In contrast, non-human parameters include wildlife grass species preferences in unprotected rangelands and natural disasters such as droughts. However, even though livestock production is central to the livelihood development of rural and farming communities, the implementation of research results regarding rangeland management and climatic variations is minimal, especially for the emerging livestock farmer’s sector. This is despite the essential inarguable negative impacts of imprudent rangeland management practices and, to an extent, the effect of long-term climatic variations translating into droughts. The rangeland and forage science literature, thus, consistently agree that below-average annual rainfall impacts on rangelands cannot be attributed to the single below-average rainfall season but to preceding seasons thereof (Mokhesengoane et al., 2021). Furthermore, Harper (1967) highlighted that in ecosystems with a recurrence of natural hazards such as droughts, populations and farmers are likelier to spend most of their 3 time recovering from the hazard. The economic impact of droughts on productivity at the farm level leads to drastically reduced returns on investment, which poses a severe threat to the development of emerging livestock farmers’ livelihoods (Van der Westhuizen et al., 2020). In recent years, this reality has become evident on grassland biome extensive livestock farmers, including semi-arid rangelands livestock farmers in the Bloemfontein Magisterial area in the central Free State province of South Africa. Benkenstein (2017) highlighted that the recent droughts, which lasted for almost a decade in South Africa, are comparable in intensity with the most devastating catastrophic droughts human nature has experienced in the past fifty years. This predicament requires good rangeland management practices from extensive livestock farmers to effectively support natural adaptation of the semi-arid rangeland vegetation to maintain sustainable rangeland productivity. Thus, Iglesias et al. (2016) indicated that the continued vulnerability of poorly managed natural vegetation for extensive livestock grazing systems in semi-arid regions leads to negative socio-economic impacts. In addition, Fereja (2017) highlighted that increased pressure on natural resources due to the exponential human population growth and livestock numbers, coupled with poor rangeland management practices, worsens the intensity of land degradation. On the other hand, urbanization exacerbated by increased shelter needs for human beings as a result of population growth, expansion of cultivation fields and excessive defoliation, as well as livestock trampling impact, leads to the inability of rangelands to withstand the extreme climatic variations and natural hazards. The world’s exponential population growth is also emphasized by the United Nations (UN) as cited by the Food and Agriculture Organization (FAO) (2014), anticipating a sharp positive trajectory on global population growth from 7.2 billion to 9.3 billion in the year 2050. Furthermore, the latest Statistics South Africa Report confirms a population growth from approximately 54 million to 62 million people. In this regard, the importance of the Sustainable Development Goals (SDGs) document, which advocates for zero hunger and sustainable livelihoods through decent job opportunities, cannot be over-emphasized. Agriculture, particularly the livestock sector, becomes a critical role player in this regard, as it currently supports livelihoods development for millions of poor people in developing countries (Swanepoel et al., 2010), including South Africa. The sector caters for sustainable livelihood development and sustenance through its daily economic activities, ranging from primary livestock 4 production, livestock transportation in between livestock farms and to different livestock marketing destinations such as livestock auctions, feedlots and abattoirs both nationally and internationally (Chaminuka et al., 2014). During climate change, the agricultural sector must increase productivity to support the increasing dietary demands of the growing population trends sustainably. Nonetheless, it must be borne in mind that prudent rangeland management practices are paramount in ensuring sustainable and increased livestock productivity during this unprecedented climate change era. This will counteract climatic variation impacts leading to droughts with discrete developmental stages in nature, which are characterised by non-tangible structural evidence. The phenomenon is, thus, exacerbated in intensity by the frequency of incidences, which too often render the rural extensive livestock farming communities incapable of recovering from the devastating catastrophic impacts. Thus, Peshawar et al. (2017) reported that the re- occurrence of below-average rainfall incidents exposes farming communities to the irreversible risk of livelihood losses. Furthermore, Mathbout et al. (2018) confirmed that below-average rainfall and poor rangeland management practices have catastrophic consequences on agricultural productivity, ecological productivity, and cultural, social, and economic sustainability in many climatic regions dependent on rainfall. Implementing rangeland management practices that ensure the improvement of rangelands, mitigate the negative effects of droughts and increase productivity essential for the livestock industry. According to a study by Van der Westhuizen et al. (2020), livestock production increased during drought season with sustainable rangeland management practices compared to no rangeland management practices during a regular rainfall season. According to Jones et al. (1990) and Davis-Reddy and Vincent (2017), the drought phenomenon is defined as a continuous period of extremely dry and hot weather characterised by a poor precipitation rate (Dai, 2011). Wardeh (1999) and Hao and Singh (2015), defined drought as a temporary recurrent phenomenon or an extended rainfall deficiency period in a geographical region. This yearly or seasonal dry spell phenomenon, influenced by climatic variation in arid or semi-arid zones, often negatively impacts poorly managed natural rangelands and livestock productivity. As the quality of rangeland resources and forage declines, livestock's vulnerability to diseases increases tremendously, leading to high livestock mortalities 5 (Mokhesengoane et al., 2021). Other scholars, such as Harrison and Shackleton (1999), have identified the African grasslands as highly resilient due to their ability to recover post-exposure to intensive grazing. Hence, in rangeland and forage science, resilient grasslands can recover and function efficiently by providing fodder, regardless of disturbances (Abel & Langston, 2001). Nonetheless, the opposing argument presented by Danckwerts and Stuart-Hill (1988), stated that the rapid recovery ability of semi-arid grasslands highly depends on sufficient grazing withdrawal periods post below-average rainfall reoccurrences. In addition, O’Connor (1995) highlights that the recovery ability of rangelands is highly dependent on the history of natural rangeland utilization, including grazing management practices employed before the onset of drought and during the drought period. According to Bosch and Tainton (1988), the Karoo, Grasslands and Savanna biomes cover eighty-five per cent of the land surfaces in South Africa. As a result, they are the most essential areas for livestock production systems. Nonetheless, the grassland biome remains the most significant part of South Africa’s beef-producing areas (Van Niekerk, 1996). The land cover type of the grasslands gives the grassland biome the advantage of being more suitable for livestock farming with specific reference to cattle and sheep compared to other biomes. However, Avenant (2019) highlighted that one- third of a grassland biome has already been transformed into other land uses such as cultivation and mining. Future rangeland degradations in the grassland biome as a result of imprudent grazing management practices, coupled with the catastrophic impact of the inevitable future drought occurrences, can profoundly impede the productive ability of the grassland biome. According to Vogel (1994) and Hoffmann et al. (2001), the research literature, government policies and legislation have pointed out the relationship between climatic variations, rangeland management and desertification. Irrespective of the latter, the South African government is still reluctant to invest money in climate change impacts research, resulting in limited knowledge in the management of climate variation impacts. Furthermore, Avenant (2019) reported that the grassland biome of South Africa covers 32 534 079 ha (Hectares), of which available livestock grazing area is 21 560 559 ha grazed by 3 934 838 LSU heads. Meanwhile, in the Free State province, the grassland area amounts to 12 982 516 ha with an available grazing area of 8 538 734 ha grazed by 1 333 815 LSU heads. Despite the latter, the grassland biome 6 remains more prone to the threat of rangeland degradation due to unsustainable rangeland management practices, including overgrazing and not having sufficient resting periods. Natural disasters, especially droughts and inappropriate land use practices, may exacerbate rangeland degradation. Nonetheless, there is an increasing research interest by the scientific community on rangelands due to their ability to sequester carbon, as they cover approximately 30% of the ice-free global land surface (FAO, 2009). The latter seeks to reduce the rapidly disruptive greenhouse gases such as carbon dioxide CO2, methane CH4, and nitrous oxide N2O concentration in the atmosphere, leading to global warming. According to IPCC (2007,2013) and MacCarthy and Zougmore (2018), the major contributors to increased atmospheric greenhouse gas concentrations are pastoralism, deforestation, crop farming, land use change, and industrial developments. However, according to Mgalula et al. (2021), restoration of degraded agricultural lands and improved rangeland management practices can significantly enhance carbon sequestration, which to an extent confirms the crucial contribution that can be made by good rangeland management practices in reducing greenhouse gas concentration in the atmosphere and related global warming risks. Rangelands are mainly utilized for livestock grazing purposes. Thus, if prudent grazing management practices are correctly followed, they can assist the root systems of grasses and shrubs to increase carbon C storage in the soil (Ma & Coppock, 2012). Over and above rangeland scientists recommend that to ensure efficient and sustainable utilization of rangelands, stocking rates must always be below or at least equal to the grazing capacity of the natural veld when rangeland is in pristine condition. However, the support from the department of agriculture and advisory services is central to land reform farmers’ development. According to Bembridge (1990), agricultural extension is a crucial field addressing many social and economic farming- related aspects in rural communities. Furthermore, Zwane (2012) confirmed that agricultural extension, as a profession, has a role to play in rural development. However, despite all these, a study by Conrandie (2016) reported a knowledge deficit of 30% and 35%, respectively, in livestock husbandry and rangeland management on newly appointed agricultural extension professionals. These manifestations threaten the sustainability of livestock productivity, mostly on farms dependent on agricultural 7 extension and related support services for guidance, namely extensive land-reform livestock farms, communal farms and commonages. 1.2 BACKGROUND OF THE STUDY The inability of emerging livestock farmers to implement sustainable rangeland management practices leads to rangeland degradation and subsequent ripple effects on emerging livestock farmers' sustainable livelihood development. On the other hand, long-term impacts of climate variations defined as droughts in successive dry years (Bounejmate et al., 2004), combined with poor rangeland management, increasing fodder shortages and rangeland degradation intensity. According to Ndandani (2016), approximately twenty-five per cent of land owned by government and rural communities in South Africa is degraded. It is all the latter factors and urbanization due to population growth coupled with the reoccurrence of the inevitable El Niño/La Niño induced climatic variations of the recent years in South Africa, Free State Province and Bloemfontein geographical region in particular, that to an extent aggravates the inability of degraded rangelands to support sustainable extensive livestock farming activities and subsequently food security. Maré and Willemse (2016) recorded an enormous catastrophic loss of 40,000 heads of livestock in Kwazulu-Natal due to poor rangelands management and below-average rainfall-induced challenges. At the same time, Mokhesengoane (2020) recorded an average devastating loss of 142.2 LSUs (Large Stock Units) amongst land reform farmers in the Bloemfontein area from October 2018 to January 2019. These mortalities increase drastically as farmers overgraze (Mokhesengoane et al., 2021). Poor rangeland management practices and below-average annual rainfall drastically reduce the ability of rangeland productivity and subsequently lead to increased susceptibility of livestock to diseases, which leads to low reproductive ability caused by a deficit in livestock dietary needs. Excessive imprudent grazing practices and consecutive years of below-average rainfall, as observed by the researcher over the years of experience in the study area, affect the ability of rangeland plants to regrow the leafy area, which is crucial for photosynthesis and related processes leading to gradual change in rangeland composition. Herbivory, below-average rainfall and poor rangeland management 8 decisions have profound underlying effects on rangeland composition and vegetation production (O’Connor, 1995). Thus, Naidoo et al. (2013) highlighted the immediate need for Southern African Development Community (SADC) region countries to prioritize rangeland management practices, which will mitigate the impacts of climate variability and, therefore, support rangelands in this region to be resilient against climatic variations. However, with sustainable rangeland practises, Van der Westhuizen et al. (2020) demonstrated that the impact of climatic variability could not only be mitigated, but livestock production levels could also be high during periods of below-average rainfall when good rangeland management practices are applied. During 2014/2015, the Southern African regions were affected by the El Niño event, and the impact became intense in 2015/2016 (Benkenstein, 2017). According to Rembold et al. (2016), the El Niño impact became evident when the first summer rains were delayed for several months. The rainfall remained significantly below average, with extremely high summer temperatures recorded. Figure 1.1: Bloemfontein rainfall differences report for hundred years from 1 October to 28 February (South African Weather Services (SAWS) long term data) 1.3 RESEARCH MOTIVATION According to Bodner and Robles (2017), rangeland and forage scientists worldwide have highlighted several rapid rangeland species compositional changes in the 9 significant grassland biomes. These changes are attributed to the combined effect of intensive defoliation by livestock and wildlife, fire and below-average annual rainfall incidents leading to invasion by exotic species and shrub encroachment. Hence, Vetter (2005) emphasized that the sustainability of rangelands in Africa and other regions is an excellent concern in rangeland management dialogues. However, O’Connor et al. (2010) indicated that a research review in South Africa failed to find a study that specifically investigated livestock management variables on plant species composition. Nonetheless, O’Connor et al. (2011) researched these variables in the sourveld of the grassland biome in Kwazulu Natal province, and their findings were that the increase in the proportion of sheep to cattle in the rangelands led to a long term decrease in forbs richness and subsequently the reduction of total species richness in general. The recent occurrence of El Niño incidents in 2014-2016, experienced across South African summer rainfall geographical regions worsened by poor management practices on rangelands, had catastrophic impacts on most of the country's rangeland resources (Baundoin et al., 2017). There is also further confirmation that below average annual rainfall impacts exacerbated by poor rangeland management practices were not only responsible for high wildlife and livestock mortalities in South Africa. Thus, they were equally responsible for the noticeable changes in rangeland species composition (Baudoin et al., 2017; Swemmer et al., 2018). Nevertheless, grass species diversity loss on the herbaceous layer is the worst impact of persistent imprudent animal stocking rates, poor rangelands management decisions and below-average annual rainfall on the grasslands. Overgrazing and underutilization of grazeable natural vegetation are significant rangeland degradation drivers in the grasslands (Du Toit & Cumming, 1999). The South African arid and semi-arid rangelands are susceptible to re-occurrence of drought incidents (Vetter, 2009). With the climate change phenomenon, drought impacts are more likely to create a favourable environment for rangelands with severe species composition loss in the next decade. Human factors such as rangeland management impact environmental pollution based on natural vegetation assessment research, which becomes critical (Zhou et al., 2018). According to Kauffman et al. (1997), rangeland degradation impacts biotic sustainability badly and reduces the various future utilizations of rangeland 10 ecosystems. Degradation of rangelands is “the retrogression of vegetative cover leading to surface layer exposure to wind and soil erosion by washing away the organic compositions that give vigour to plants’ development” (Solomon, 2003). Thus, Fynn and O’Connor (2000) highlighted that intensive grazing during droughts substantially declines perennial grass species such as Themeda triandra in the semi-arid grasslands. On the other hand, there is limited research highlighting the relationship between rangeland condition, soil chemical properties and soil pH levels, apart from the findings of a study done by Kotzé (2015) in the eastern part of Bloemfontein which investigated the concentration of these soil chemical properties at different soil depths under different rangeland management systems. Due to the world population growth and the shift from nomadic livestock farming to a more relaxed livestock farming fixed to an area, Squires (2010) indicated that many land-use methodologies and strategies are irrelevant due to new economic and political trends. Furthermore, Grobler et al. (2014) suggested that the South African livestock sector’s self-sufficiency can be achieved by increasing the average calving rate and the beef production off-take, mainly in the communal and emerging sectors. The latter will be impossible without prudent and sound rangeland management practices, which ensure both production efficiency and sustainability of the rangeland’s ecosystem and, to an extent, mitigation of catastrophic below-average rainfall impacts on rangelands. 1.4 RESEARCH AIM To provide holistic guidance for future sustainable livestock production in the semi- arid rangelands of Bloemfontein magisterial area for land-reform farmers. 1.5 RESEARCH OBJECTIVES 1. To investigate land-reform livestock farmers’ decisions on management to establish their production efficiency. 2. To compare rangeland condition per land-reform farm and between land-reform farms. 11 3. To establish correlations between rangeland condition, soil chemical properties and managerial inputs on the sampled land-reform farms. 4. To establish a correlation between rangeland condition and soil carbon content. 1.6 RESEARCH QUESTIONS 1. What holistic management guidance can be provided to extensive land-reform livestock farmers in the semi-arid rangelands, which will lead to sustainable livestock productivity during this unprecedented climate change era? 2. Does extensive land-reform livestock farmers' management decisions influence their production efficiency? 3. What are the similarities between rangeland conditions per farm and between different land-reform farms? 4. Does a correlation exist between rangeland's condition and soil chemical properties? 5. Does a correlation exist between soil carbon and rangeland conditions? 1.7 HYPOTHESIS This study hypothesizes that poor rangeland management implicates rangeland degradation with lower condition scores, influencing soils and production levels and increasing the negative impacts of drought and global warming. 12 CHAPTER 2 LITERATURE REVIEW 2.1 INTRODUCTION According to FAO (2013), the grasslands cover an estimated surface of seventy per cent of the global agricultural area, which accounts for twenty-six per cent of the total global land area. Nonetheless, the world’s rangelands are subjected to active management practices based on several methods and challenges Snyman (2009), and the South African grassland biome is no exception, particularly the Bloemfontein area in the central Free State. Research in the development of agro-ecological rangeland condition assessment techniques has always been a priority for rangeland scientists in the last sixty years (Bosch & Taiton, 1988); to monitor rangeland condition and to quantify forage production, the rangelands can offer livestock grazing. Over the past years, techniques have focused on benchmarks, ecological indexes, key species, and degradation of gradient rangeland condition assessment methods (Friedel, 1991). Finally, in 2003, Van der Westhuizen (2003) developed an objective rangeland degradation gradient method for the semi-arid rangelands of the central Free State province region in the grassland biome of South Africa. 2.2 RANGELANDS MANAGEMENT According to Tefera et al. (2010), communal and commercial rangeland management systems are the most common systems utilized for extensive pastoral farming. However, the natural rangeland production hindrances, such as poor rangeland management, namely, overstocking and understocking, as well as water balance and nutrient availability, were identified in these systems. Thus, the ecological sensitivity of semi-arid rangelands continuously increases their susceptibility to excessive grazing pressures, leading to rapid deterioration of these rangelands if not correctly managed (Van der Westhuizen et al., 1999). Despite this, SANBI (2014) discovered that pastoralists often manage livestock with the intent to achieve production goals only and not to maintain the rangeland's 13 ecosystem. This is why Van der Westhuizen et al. (2018) reiterated that optimization of long-term forage production quality needs rangeland deterioration prevention through good rangeland management practices. According to Bosch and Taiton (1988), rangelands utilization history and climatic variation information are very crucial for rangeland management planning, rangeland improvement, animal species choice and the number of animals the rangeland can support since the livestock production systems differ according to vegetation nature. Solomon et al. (2006) also reported that climatic variations coupled with improper grazing adversely affect the composition and quantity of grasses in the seed bank. In summary, Van Der Westhuizen et al. (2018) classified South African extensive grazing management into three categories: communal production, with no land ownership and poor farming infrastructure maintenance. Thus, continuous grazing is commonly used. The authors described the second grazing management category as primarily used by emerging livestock farmers, who graduated from communal farms and varies from constant grazing to two, three, and four-camp systems, with visible signs of rangelands degradation. The third category is commercial production, characterised by well-maintained critical livestock farming infrastructure (water reticulation and well-fenced-off and divided grazing camps). As a result, a multi grazing camp system is used. Nevertheless, according to Allsopp et al. (2007), the intentions to promote ethical livestock farming and rangeland management practices amongst smallholder emerging and communal farmers have failed drastically in Africa. This is partly because of non-compliance to ecological carrying capacities in arid and semi-arid regions. Nonetheless, the South African agricultural sector originated from a demise that agricultural input support programs initiated by the government were mainly intended for large-scale and commercial farmers focussing on productivity and contributing to the gross agriculture income. Thus extending the minimum to no support at all for smallholder emerging farmers and communal farmers (Vetter, 2003). Despite this, the researcher, over the years of experience, has observed that agricultural support extended to land-reform and communal farmers by the current government during natural disasters such as droughts is in contrast with ethical rangeland management 14 practices, since farmers with more livestock numbers receive more assistance as opposed to those with fewer livestock numbers, without careful considerations of farmers grazing area sizes. However, too often than not, as highlighted by Salomon (2011), these efforts ignore sustainable rangeland utilization goals and sustainable livestock productivity goals by farmers, jeopardizing future livestock reproductive performance and quality of weaner animals produced from extensive livestock farming systems. 2.3 GRASSLANDS VEGETATION The Grassland biome of South Africa, as a significant resource for livestock production, covers parts of the Free State, Kwazulu Natal, Eastern Cape, Mpumalanga Provinces and parts of the Mountain Kingdom of Lesotho (LandCover 2013/14 Data Set, 2015). Figure 2.1: South African grasslands biome map (LandCover 2013/14 Data Set, 2015) Vast differences in annual average rainfall characterise these grasslands. As a result, they are dominated by sourveld grass types in the higher annual rainfall areas and sweetveld grass types in the lower annual rainfall areas (Naidoo et al., 2013). The dominant grass species used as rangeland condition indicators are Themeda triadra (climax) for good rangeland condition score, Eragrostis (sub-climax) species for 15 average rangeland condition score as well as Aristida species and Cynodon dactylon (pioneers) for abysmal rangeland condition score (Van der Westhuizen, 2003). Natural vegetation for livestock grazing in Bloemfontein is described as sweet grass veld of the grassland biome, with Themeda triandra as the most distinctive and well- distributed grass species, followed by the Eragrostis grass species (Acocks, 1988; Van der Westhuizen, 2003). Trees are extremely limited in plain topography variations, but dense stands can occur in vlei areas and hill variations. Prudent veld management is essential to maintain the rangeland condition on an optimal positive trajectory to ensure sustainable livestock production. The distinctive characteristic of Bloemfontein rangeland sweet veld type of grasslands, with that of the sour veld type of the grassland biome, is low annual average rainfall leading to constant palatability with low carrying capacity, when compared to the sour veld with decreasing palatability as it grows with high carrying capacity, due to higher annual average rainfall (Tainton, 1999). Nonetheless, “the grassland biome remains the greater part of South Africa’s beef- producing areas as highlighted” (Van Niekerk, 1996). This is despite the observations by Musemwa et al. (2012), who suggested that livestock farming is not the key enterprise in the emerging and communal farming categories compared to other agricultural enterprises in different biomes. However, the suitability of the grassland biome for extensive livestock farming activities is inarguable. 2.4 LIVESTOCK REPRODUCTION PERFORMANCE Over the years of experience, the researcher observed that Bloemfontein's extensive land-reform livestock beneficiary farmers mostly come from communal livestock farming systems, which are characterised by poor application of grazing management systems, high land degradation incidents and low calving rates of 35% and below as reported in the South African livestock structured survey conducted by (Scholtz & Bester, 2010). Furthermore, Nkadimeng et al. (2022) reported the average reproductive performance of smallholder cattle farmers in South Africa as 50% pregnancy rate less 12% of losses, indicating an average calving rate of 38%. These low calving rates are comparable to the average 32% calving rate of land-reform livestock farmers in Bloemfontein in the central Free State province of South Africa, 16 reported during the 2018/2019 drought in a stocking rate study conducted by (Mokhesengoane et al., 2021). The latter emanates from high stocking rates and ineffective rangeland management, which yields poor reproductive performance, low growth rates and take-off (quantity of meat produced over time) due to the lengthy periods it will take for animals to be marketable. Thus, these low calving rates are also comparable to the results of a research trial conducted by Van der Westhuizen et al. (2020) at the Glen experimental farm in Bloemfontein, which recorded a 32% calving rate with minimal to no rangeland management effort during drought. These low calving rates negatively reflect on the livestock production efficiency of the Bloemfontein land-reform farmers' group, threatening the sustainability of their livestock farming operations. Nevertheless, these low calving rates can be attributed to a range of interlinked livestock production factors: animal nutrition, animal breeding and animal disease control, with management as a core factor or at the centre of sustainable livestock productivity. However, sustainable livestock production management must also include ecological sustainability by preventing rangeland degradation under variable climatic conditions (Dankwerts & Tainton, 1996; Van der Westhuizen et al., 2018). 2.5 LAND DEGRADATION According to Mani et al. (2021), land degradation is a worldwide challenge with adverse effects on ecosystem functioning and livelihood development. However, the total size estimates of degraded lands worldwide are uncertain (Gibbs and Salmon, 2015). In South Africa, land degradation is rife, which was confirmed by approximately 60% of degraded lands (UN Environment Programme, 1997). Thus, in their research projects, Stocking and Murnaghan (2001) have identified inappropriate land management, wrong technology, exponential population growth, poverty and decisions of social and political formations as key human factors related to land degradation. Nonetheless, other studies presented a different viewpoint by highlighting that land degradation can occur independently from human activities but worsen with poor rangeland management practices. It was further confirmed by 17 Darkoh (2009) that land degradation and overgrazing are considered among the key environmental challenges in South Africa. On the other hand, Belsky and Blumenthal (1997) indicated that livestock activities, namely grazing and trampling, affect plant species composition directly, even though the severity of the impacts differs according to animal distribution and density. Since plant species vary significantly in their palatability and response to grazing pressures, species composition can be used to indicate rangeland conditions (Abule et al., 2007). According to Oztas et al. (2003) and Maki et al. (2007), changes in vegetation structure, composition and productivity occur due to changes in grazing pressure. Increased livestock grazing pressure increases palatable plant species' susceptibility to heavy grazing and livestock trampling impact. The latter is evident by bare patches near water troughs and livestock feeding points, leading to the disappearance of Decreaser plant species as they are replaced by Increaser or Invader plant species as reported by (Sisay & Baars, 2002). Thus, Abule et al. (2007) indicated that biomass production can be utilized to determine the quantity of available forage for animal grazing. 2.6 CLIMATE VARIATION The South African grassland biome’s climate is typical of the sub-tropical semi-arid environment, which is characterized by sweltering summers and cold dry winters, with the mean minimum temperatures ranging from minus nine degrees Celsius in winter to the mean maximum of thirty-nine degrees Celsius in summer (J. Van den Berg, personal communication, August 29, 2022) Annual average rainfall in Bloemfontein is 556 mm and the substantial amount of this rains fall during summer months (ISCW- databank, 2018). Hence, the Grassland biome is generally classified as an arid to semi-arid region receiving an average rainfall of 500 mm or less per annum, which applies to approximately 65% of the grasslands (Snyman, 1998). Due to low annual average rainfall and high temperatures, the grasslands are more prone to periodic climatic variations. 18 Nevertheless, rainfall has apparent effects on herbaceous productivity, and as a result, rainfall timing is essential for rangeland recovery post-defoliation and exposure to variable climatic conditions. Rangeland's growth in the central Free State increases tremendously from November to February as days are longer and nights are shorter during this period. However, Fynn et al. (2000) indicated that the importance of rainfall on veld compositional change cannot be attributed to the single rainfall season but also the preceding seasons. Bloemfontein respectively received below-average rainfall from October to February in 2016/2017, 2017/2018 and 2018/2019. As a result, livestock productivity was severely impacted. Thus, Dzama (2016) indicated that the 2015/2016 El Niño induced droughts compelled Lesotho, Swaziland, Namibia, Malawi and Zimbabwe to declare a state of drought emergency nationally, whilst two of the SADC countries, namely South Africa and Mozambique, declared drought emergencies partially. However, the researcher also became cognizant of the fact that in a natural weather pattern environment, El Niño events precede LA Niña events, which is the total opposite as LA Niña is characterised by above-average annual rainfall leading to flooding and freezing winter temperatures (Sazib et al., 2020). Thus, Van Der Westhuizen (2006) reported that an average of approximately sixty- two per cent contribution to rangeland production occurs during these five months. However, this process heavily depends on adequate rains from August to October, ending the dormant winter period with shorter days and longer nights. Nonetheless, climatologists have already predicted a sharp increase in the climatic extremes and severity in the twenty-first century in both arid and semi-arid regions (South African Weather Services). However, Keshavarz et al. (2017) highlighted that these occurrences will severely impact rangeland's sustainability. Thus, sound and appropriate rangeland management practices will be imperative in minimizing the anticipated negative impacts. 2.7 SOIL PROPERTIES In South Africa, more emphasis is placed on the impacts of rangeland management practices on livestock performance and forage production. However, less emphasis is 19 given to the effects of grazing practices on soil properties (Snyman & Du Preez, 2005; Du Toit et al., 2009). Nevertheless, soil is the most essential resource for ecosystem functioning. It is equally crucial for both physical and chemical soil properties, which give vigour and are necessary for healthy flora production in the ecosystem. Apart from its significant role in livestock production, this ecosystem flora also prevents soil erosion and thus further prevents degradation of these ecosystems. According to Greenwood et al. (2001), above-ground vegetation loss by intensive grazing on rangelands and soil macro-pores loss tremendously increase the water runoff rate. Chen et al. (2008) also indicated that the loss of vegetation cover might lead to changes in the soil P transformation, resulting in changes in plants' P needs and soil’s physical, chemical and biological properties. The intensity will, however, depend on the soil texture form being the proportional representative of clay, sand, and silt. According to Pei et al. (2008), overgrazing of rangelands is responsible for a significant decrease in soil chemical, physical, and biological properties. Nonetheless, Derner and Schuman (2007) attested that the carbon sequestration process is critical for rangeland health maintenance and rehabilitation, as it enhances the improved soil structure, nutrient cycling, and general soil quality and ultimately reduces soil erosion since it increases soil water retention capacity and thus limit water run-off. Kotzé (2015) provided empirical evidence showing that grazing directly impacts soil properties such as organic C, extractable P, exchangeable Ca and Mg, in research trials conducted in the grasslands and the savannah biome, respectively. The study furthermore highlighted that some soil properties are more sensitive to rangeland condition degradation than others (e.g. POM and soil aggregates). These soil properties are also essential for the resilience of rangelands to degradation. The trampling impact of livestock on rangelands exacerbated by overgrazing leads to the removal of vegetation cover, thus rendering the soil properties susceptible to being washed away during heavy rains. Thus, Fernandez-Giminez and Allen-Diaz (2001) confirmed that most changes in soil properties and nutrients occur in the livestock watering points and supplementary feeding areas due to the impact of livestock trampling. 20 2.8 POLICY Greenberg (2010) criticised the South African national government for failing to prioritise agriculture since the advent of democracy, citing a low budget of below two per cent of the national budget allocation to national and provincial agriculture departments even though significant but not satisfactory budget allocation adjustments have been made to date. This ignites serious concerns, especially when the world will be ushering in a new era of climate change, in which scientists have already predicted inevitably above-average rainfall patterns and above-maximum average temperatures, translating to more floods and drought incidents re-occurrences. Central to the latter situation is the exponential growth in the human population, as highlighted by forever rising food and fuel prices, abrupt changes in the political arena, and the absence of clear policy direction to unlock land reform challenges, which contributes immensely to the underdevelopment of drought and climate change impacts mitigation strategies in the South African arid and semi-arid rangelands (Vetter, 2009). Thus, Van der Westhuizen et al. (2018) indicated that successive droughts reduce perennial plant cover and subsequently decrease the number of animals the natural veld can support under normal circumstances. Despite the regulatory framework provided by the policy, which seeks to create an enabling environment by steering the government investment such as human resources and finance for implementation and research priorities articulation as highlighted by Lahiff (2006) and Vetter’s (2013) argument is aligned to that of Hall and Clife (2009) highlighting that government’s targets for agricultural land redistribution and to improve both food security and rural livelihoods have not been realised despite land-reform programs enjoying political support. 2.9 SOUTH AFRICAN LAND-REFORM HISTORY In South Africa, Land reform aspired to address more than 350 years of race–based colonization and dispossession as part of the transition agenda to a democratic society (Olive et al., 2017). This programme in South Africa emanates from the election manifesto of the African National Congress (ANC) as a build-up to the first democratic elections held in 1994, leading to the advent of democracy. The main objective of this 21 programme was to redress the injustices of the apartheid system by securing tenure, generating employment, increasing rural incomes, supplying residential and agricultural land for historically disadvantaged individuals and increasing agricultural productivity. As a result based on the mandate in the constitution land reform in South Africa has three pillars, namely; restitution, tenure reform, and redistribution. The pillars involve returning land or providing other compensation to persons or communities who were dispossessed of property after 1913 due to racially discriminatory laws or practices (ANC, 1994). The techniques employed by the government to realize land redistribution were based mainly on the operation of the existing land market. Amongst other measures the government can explore is land expropriation without compensation, even though it has not been utilized widely thus far. South African government’s role has always been limited to support through provisional grants to assist people who cannot enter the land market to buy their own property. The willing seller –willing buyer (WSWB) concept slowly entered the discourse in the South African land reform issue during 1993-1996 as a reflection of the ANC’s fast shift in economic thinking from the nationalist agenda to the neoliberal agenda (Lahiff, 2006). Exhaustive, extensive consultative efforts by the new Department of Land Affairs (DLA) within South Africa and international advisers pointed to a new policy direction, which was outlined in the 1997 White paper on South African Land Policy, leading to market-based approach and particularly the WSWB concept as the cornerstone of land reform policy (World Bank 1994; Williams 1996; Hall et al., 2003;). The South African Constitution did not guide the approach. Still, it was a policy choice to keep up with emerging international trends and the macroeconomic strategy (the Growth, Employment, and Redistribution strategy) adopted by the ANC in 1996. In 2000, the Settlement /Land Acquisition Grant (SLAG) was rolled out, prioritizing citizens living in abject poverty with a maximum monthly household income of R1, 500 to qualify for the R16,000 grant. Even though success was realized in this regard, it was widely criticized for clustering too many poor people on one commercial farm without the necessary skills and resources to make those farms productive. The SLAG program was then effectively replaced by a program referred to as Land Redistribution for Agricultural Development (LRAD) in 2001, which was introduced with a clear objective to advance commercially focused agriculture but alleged to cater for other 22 farming groups such as commonage farmers. The new policy offered higher grants paid to individuals rather than households; it also used loans financed through institutions, such as the state-owned Land Bank, to supplement the grant. LRAD offered a single unified grant support system which allowed beneficiaries to access money on a sliding scale ranging from R20, 000 to R100, 000 (T. Kuduga, personal communication, May 20, 2023). The applicants had to bring their contribution to qualify. However, their contribution was not limited to cash since the beneficiaries’ labour was also accepted to replace monetary contribution. To qualify for the minimum grant of R20, 000 the beneficiary was supposed to contribute R 5,000 in cash or as labour. Provincial land reform offices provided the LRAD grants as per the agency contract with the DLA, and the grant was then paid out through the Land Bank. The LRAD program’s land acquisition approach retained the market-based, demand-led approach of the previous policies. As of 2006, the government became proactive and started buying land on the market without identifying potential beneficiaries prior to land purchasing. Most redistribution projects have involved groups of applicants pooling their grant allocation to buy formerly white-owned farms to pursue commercial agriculture. The focus on group projects led to mostly the small size of the available grant relative to the size and cost of the typical agricultural holding and many more challenges associated with the subdivision of land. Most of the rural communities viewed redistribution as a means to expand their existing communal landholding, and they also favoured collective ownership. According to van den Brink et al. (2007), LRAD farmers made smaller groups comprising extended family groups due to the increased availability of financing in the form of grants and credit. However, it later came to the realization of the South African government that conflicts were rife amongst these groups, and as a result, they later sold these farms back to their previous white owners since they had title deeds. Thus, the government resolved to implement the Proactive Land Acquisition Strategy (PLAS) for agricultural land redistribution. In this program, land ownership lies solemnly with the state and the beneficiaries are given thirty-year lease agreements with an option to buy. It also allows the government to prematurely terminate the contract with the beneficiary through its provincial subsidiaries if the land is not utilized fully for agricultural purposes. 23 Despite all the efforts, the production levels of most land-reform farmers in South Africa are not satisfactory (Shabangu et al., 2021). This can be attributed to the socio- economic status of land-reform beneficiaries, incomprehensible government support channelled towards land-reform farmers and the sentimental values some land-reform farmers attach to livestock. Nonetheless, according to Sikwela et al. (2018), land reform in South Africa remains critical to unlocking social and economic transformation and bringing about true reconciliation between different ethnic groups. Due to the nature of the grassland biome ecosystem, land reform farmers in the grassland biome are mostly extensive livestock farmers. Thus, rangeland management is critical for successful livestock farming in this biome. 2.10 CONCLUSION According to Van der Westhuizen (2003), Themeda triandra in the grassland biome rangelands is critical for the proper functioning of the ecosystems. For this reason and many other ecological benefits, Van der Westhuizen et al. (1999) identified Themeda triandra as the most important grass species as a rangeland condition indicator and the grassland's essential livestock diet composition. Nonetheless, proper rangeland management practices are paramount for maintaining these ecosystems and ensuring their sustainable contribution to livestock productivity. The recent reoccurrences of below-average rainfall incidents in these ecosystems, perpetuated by climate variations, is evidence enough that rangeland management is the only and most important variable a farmer or rangeland manager can control. Thus, Van der Westhuizen et al. (2020) highlighted that livestock production efficiency can be achieved with proper veld management and without production licks supplementation since many studies have already identified adverse impacts of livestock production licks supplementation on the soil chemical and physical properties (Van der Westhuizen et al., 2020; Owens et al., 2021). Over the past three to four decades, scholars have been in disagreement about the climate change reality’s impacts, intensity and anticipated effects on agricultural productivity over time, with some arguing that climate change reality was too often cited out of context, blown out of proportion, loosely used and often mistaken for 24 normal weather patterns, Benestad et al. (2015), these authors reviewed a total of thirty-eight scientific publications disputing climate change reality. The latter is appropriated to have resulted in the reluctance in agricultural policy formulation by governments, specifically rangeland protection policies. However, the government has a crucial role in ensuring agricultural policy development and creating an enabling environment for such policies and sustainability of the agricultural sector to decrease food insecurity as the population grows. Even though the South African government does assist farmers during natural disasters such as floods and droughts as per the South African Disaster Management Act 57 of 2002 (as amended, Act 16 of 2015), it is imperative to highlight that the due diligence process before the natural disaster declaration which might lead to government support through reliefs in the form of either livestock feed, medication, seeds, fertilizers and water infrastructure upgrades is lengthy and thus rangeland management and proper stocking rates are paramount at farm level to eliminate livestock losses. 25 CHAPTER 3 MATERIALS AND METHODS 3.1 STUDY AREA/LOCATION The study was conducted in the Bloemfontein magisterial area situated in the semi- arid region of South Africa, lying at an altitude of 1.395 metres above sea level. According to Acock's (1988) classification, Bloemfontein is within the grassland biome with Themeda triandra as the most distinctive and well-distributed grass species amongst other perennial grasses for livestock grazing (Van der Westuizen, 2003). Figure 3.1: Bloemfontein magisterial area map highlighting the position of studied farms (Available:https://www.google.com/maps/place/Bloemfontein) The topography of Bloemfontein is characterised by the scarcity of trees in the plain variations, with dense stands found in hill variations and shallow minor lake areas. According to Van der Westhuizen (1994) and Van der Westhuizen et al. (1999), Themeda triandra is a crucial significant species in this vegetation type; thus, the species is not only of ecological importance in the study area but also an excellent rangeland condition indicator (r2 = 0.99). Themeda triandra is the most important grass 26 species in the central Free State's extensive livestock diet composition (Van der Westhuizen et al., 2001). According to Van Weshuizen et al. (1999), Themeda triandra and Eragrostis chloromelas are dominant grass species for veld in reasonable rangeland conditions, with scores of 50% and more. In contrast, Themeda triandra dominates the veld with an excellent rangeland condition score of 90% and more. However, with degradation, Themeda triandra as climax species becomes the first grass species to be replaced by Eragrostis chloromelas as sub-climax, followed by Aristida bipartitah as pioneer grass species. Nonetheless, Themeda triandra also occurs in poor rangeland conditions with a score of 30%; however, it is not the dominant grass species. The Department of Agriculture and Rural Development (2003) recommended 6ha per LSU head animal stocking rate for the Bloemfontein semi-arid rangeland geographical region. However, as rangeland conditions deteriorate due to unsustainable rangeland management practices, grazing capacity decreases tremendously in the semi-arid region of the central Free State. Thus, rangeland managers and other key agricultural role players in this region must understand that these animal stocking rate recommendations are for the natural veld in good condition. Nonetheless, in practice, rangeland managers must interpret these recommendations with the natural forage scarcity periods and natural disaster occurrences such as droughts. However, according to Chaichi et al. (2005), the soil form and veld management practices influence grass species re-establishment and root development in the rangelands. This is why soil water retention capacity is essential for rangeland's sustainability and establishment. Nevertheless, MacVicar et al. (1977) described Bloemfontein soil form as dominated by Milkwood, Arcadia and Valsrivier, with more than 35% clay. 3.2 RESEARCH METHODOLOGY Brannen (1992) provides the following distinctive criticism for qualitative and quantitative research methodologies: quantitative methodology assumes a broader world perspective, whereas qualitative methods take a narrow view of the world. Although qualitative methodology assumes a narrow perspective, it produces an objective analysis of biological processes indispensable for research purposes. Thus, 27 the two methodologies will be employed together in this study to strike a balance between the inefficiencies of each and for a practical holistic research problem approach. Quantitative methodology was employed to question the hypothetical statement of this study by establishing the technical knowledge of rangeland managers, current stocking rates, and grazing camps' utilization history in the sampled sites. In contrast, qualitative methodology was used to test this study's hypothetical statement by establishing a basis for rangeland degradation and rangeland changes comparison on the nine sampled sites. Creswell (2013) confirmed this mixed methods approach for effective cross-checking of results by utilising different techniques to ensure a holistic research problem approach and objectivity. 3.3 RESEARCH INSTRUMENTS Structured questionnaires were used to determine profiles of rangeland managers, current stocking rates and grazing camps utilization history and management for the nine sampled farmers. The Wheel Point Apparatus Field Method by Tidmarsh and Havenga (1955) was used to identify and determine grass species composition on two managed grazing camps per farm identified by the farmer, researcher and rangeland scientist; soil samples from each sampled site were collected with a soil auger, from a 20 cm layer for comprehensive soil laboratory tests of the following soil chemical properties electrical conductivity (EC), sodium adsorption ratio (SAR), calcium (Ca), magnesium (Mg), potassium (K), sodium (Na), zinc (Zn), soil organic matter (SOM), soil carbon (Co), soil pH using potassium chloride (KCl) method instead of water (H2O), phosphorus (P) using P-Bray 1 method. To determine soil texture, three fractions, namely clay, silt and sand, were used (The Non-Affiliated Soil Analysis Work Committee, 1990). The grass species were categorized according to their classes’ Climax, sub-climax and pioneer to determine the greater class proportionality to identify the rangeland condition per grazing camp per sampled farm. The results were then used to compare rangeland conditions on the sampled camps per farm and between farms. 28 3.4 SAMPLING PROCEDURE Nine extensive land reform livestock farming sites out of a total of twenty-nine land reform farms which participated in the stocking rate study during the 2018/2019 drought were sampled using a snowball sampling procedure (Bhattacherjee, 2012), starting with the perceived group leader within the stocking rate study sampled group. Each farmer was requested to identify from a land-reform farmers group one farmer they perceive to be generally a good livestock farmer in their category. Although this technique could be biased regarding the selection of farmers with better livestock skills, it was used to allow farmers to be involved in the selection process from already scientifically selected participants. This technique is, however, also subjective in the sampling process of farmers and will probably not represent a random selection of farmers over the range of different stocking rates and management skills, therefore not representative of varying levels of farm management but instead focus on some of the best livestock farmers according to perceptions of the other Bloemfontein land- reform beneficiary farmers. Nevertheless, the risk of bias is low when snowball sampling procedure is employed on a homogeneous population. 3.5 SAMPLE Nine extensive land reform livestock farming sites out of twenty-nine land reform farms that participated in the stocking rate study during 2018/2019 drought were sampled using a snowball sampling procedure, starting with the perceived group leader within the sampled group. Each farmer was requested to identify from a land-reform farmers group one farmer they perceive to be generally a good livestock farmer in their category. The snowball sampling procedure was used to eliminate the unforeseen biases from the researcher's side and ensure objectivity in the study's outcome. Financial implications informed the sample size for this study since the researcher financed the data collection process. 3.6 DATA COLLECTION Structured questionnaires (see appendix A) were completed in an interview with each sampled site rangeland manager/farmer to establish their rangeland management, technical knowledge, literacy level and demographic information. 29 During the interview, the researcher gave a detailed explanation of the exercise to the farmers. The researcher then poses questions as they are captured on the structured questionnaire and records the rangeland manager’s responses. The narrative data captured contributes to the achievement of the main objective of this study. During September and October 2021, the agro-ecological veld condition assessment was conducted in all nine sampled farms, using the wheel point apparatus method, as confirmed to present an acceptable level of accuracy and repeatability by (Walker, 1970; Sykes et al., 1983). During agro-ecological veld condition assessment, abundance data were collected by recording the nearest plant for 100 points per camp. A maximum of 100 points were taken per two camps of all sampled farms to determine rangeland condition percentage scores (Frequency = No of plants/ total count x 100) using the objective rangeland degradation gradient method developed by (Van der Westhuizen, 2003) for the semi-arid rangelands of the central Free State province region in the grassland biome of South Africa. Big bare patches around watering points and feeding areas were purposely avoided. Independent soil samples were taken from each sampled camp per farm at the beginning of the rangeland condition assessment transect line, in the middle and end, for soil chemical and physical properties analysis. Soil auger was utilized to take soil samples at 20cm deep. Thus, Van der Westhuizen et al. (2005) emphasised that a periodical objective assessment of rangeland conditions is essential to monitor rangeland condition trends. Furthermore, Foster (2015) further confirmed that the success rate and reliability of rangeland condition assessment results rely entirely on the experience and vegetation knowledge of the individual undertaking the assessment. Thus, the researcher was closely supervised during the ecological rangeland condition assessment on each sampled farm for the reliability of the data collected by Dr H.C Van der Westhuizen, a Professional Rangeland and Forage scientist with more than thirty years of practical experience in this field (see Appendix B). 3.7 DATA ANALYSIS Statistical analyses were performed on all variables. The descriptive statistics, means and standard deviations were calculated for comparison purposes, and analysis of 30 variance (ANOVA) was also used to compare the average values. Metabolic body weights were used to calculate large stock units (LSU) (Meissner et al., 1983). Rangeland conditions were determined for every camp using the indicator species technique developed and tested against a degradation gradient technique specifically for this vegetation type (Van der Westhuizen, 2003). All soil chemical analyses were performed and done according to (The Non-Affiliated Soil Analysis Work Committee 1990): soil pH was measured using potassium chloride (KCl) and plant-available P using P-Bray 1 method (The Non-Affiliated Soil Analysis Work Committee 1990). The quantitative data was analysed using the Statistical Package for Social Science (SPSS) version 28 for descriptive analysis. 3.8 ETHICAL CONSIDERATIONS The researcher considered the ethical research conduct protocol of the University of the Free State in conducting this research by seeking and obtaining ethical clearance from the university’s ethical clearance committee on general human ethics protocol (UFS-HSD2021/1018/21) and environmental and biosafety research ethics committee (UFS-ESD2021/0123/01) respectively, prior data collection (see appendix C). The ethical clearances are attached to the appendix list of this research report. The researcher further stated that the data would be used discretely without divulging respondent information. Thus, the data cannot be traced to a farmer or farm. The data will also be stored following each ethical clearance committee's data storage requirements. Furthermore, the study followed the required academic standards by properly citing all the materials used in the study and the material authors in the reference list. 31 CHAPTER 4 RESULTS AND DISCUSSIONS 4.1 PARTICIPANTS OVERVIEW The sampled extensive livestock farms are approximately 50km around Bloemfontein city, as shown in Figure 3.1. The farm owners’ literacy level differs significantly, ranging from Standard 4, the minimum highest standard passed, to a university master’s degree, the highest qualification obtained. Nonetheless, neither of the farmers had a formal qualification in the agricultural field. The average age of the sampled farmers was 53 years, comparable to the findings of Mokhesengoane (2020). More than half of the group regarding gender representation were men, with only two women and one person aged less than 35 years. Thus, this sample group is consistent with the general perception that the South African agricultural sector is male- dominated. Furthermore, it correlates with the findings of Anyanwu (1992) that younger South African men have little to no interest in the agricultural sector. For this reason, most of them, from farms or rural communities, go to big cities nationwide to look for glamorous lives and jobs to sustain their livelihoods. All the sampled farms have been active extensive livestock farms under the same management for over a decade, and all have extensive commercial livestock farming history before 1994. 4.2 SPECIFIC INDIVIDUAL PARTICIPANT NARRATIVE 4.2.1 Participant 1 The farmer had a total farm size of 2484 ha, of which the total portion available for livestock grazing was 1880 ha. The total Large Stock Units kept for extensive farming purposes were 157.81 LSUs. Rangeland condition scores recorded for this farm post rangeland condition assessment were 62% and 61%, respectively, with the average rangeland condition score of 61.5%. The farmer had no grazing management system for effective rangeland management. Themeda triandra was the most prevalent grass species identified during the rangeland condition assessment, followed by the Eragrostis chloromelas. The total calving rate of 28% and the average weaning weight 32 of 192kg at roughly six to seven months of age were recorded on this farm. The farmer followed a prescribed animal health program for the area but did not follow a breeding season and had no production plan. The farmer supplied seasonal production licks to livestock as well as salt blocks. The highest education level recorded for the farmer was matric, and the farmer’s trusted agricultural information source was only the informal mentorship arrangement with the neighbouring farmers. 4.2.1.1 Production level A low yearly calving rate of 28% was recorded, with pre-weaning annual mortality of 15 calves and yearly mortality of 18 LSUs recorded on matured livestock. The average weight of weaners recorded over time was 192kg, and this farm recorded an exceptionally low stocking rate of almost below 50%, resulting in an additional substantial loss of income. 4.2.1.2 Financial implications Under- and overstocking is neither beneficial to the ecological sustainability nor the profitability of extensive livestock farming enterprises (Mokhesengoane et al., 2021). Thus, based on the beef average weaner price of October 2021 of R39.50 (see appendix E) and the average class C2/C3 beef carcass price of R45.11/kg, the farmer lost R 113 760 of the potential weaner income and R 194 875.2, respectively on matured cow slaughter income. 4.2.2 Participant 2 The farmer had a total farm size of 516 ha, of which the total portion available for livestock grazing was 376 ha. The total Large Stock Units kept for extensive farming were 66.19 LSUs. Rangeland condition scores recorded for this farm post rangeland condition assessment were 19% and 56%, respectively, with the average rangeland condition score of 38%. The big difference in rangeland condition is because one survey was executed on old arable lands. The farmer, however, identified this camp because it was regularly grazed. The farmer was using a rotational grazing system for effective rangeland management. Eragrostis chloromelas was the most prevalent 33 grass species identified during the rangeland condition assessment, followed by Themeda triandra and annual herbs on the old arable lands. The total calving rate of 85% and the average weaning weight of 186kg at roughly six to seven months of age were recorded on this farm. The farmer did not follow a prescribed animal health program and the breeding season for the area but had a production plan. The farmer supplied no seasonal production licks to LSUs kept on the farm but instead, provided salt blocks. The highest education level recorded for the farmer was matric, and the farmer’s trusted agricultural information sources were internet, government agricultural extension and private agricultural extension support. Stock theft is a significant problem in this area, and the farmer kraals livestock every evening. 4.2.2.1 Production level A high yearly calving rate of 85% was recorded, while no yearly pre-weaning mortality was recorded on this farm, with mortality of 2 LSUs recorded for matured livestock only. The average weight of weaners recorded over time was 186kg. The insignificant stocking rate, compared to the area carrying capacity norm of 6ha/LSU, was also recorded on this farm. 4.2.2.2 Financial implications Proper stocking rates are essential for the sustainability of rangeland ecosystems and efficient livestock productivity. Thus, based on the average class C2/C3 beef carcass price of R 45.11 in the central Free State. The farmer lost R 21 652.8 of potential matured cow slaughter income. 4.2.3 Participant 3 The farmer had a total farm size of 260 ha, of which the total portion available for livestock grazing was 252 ha. The total Large Stock Units kept for extensive farming purposes were 83.48 LSUs. Rangeland condition scores recorded for this farm post rangeland condition assessment were 49% and 54%, respectively, with the average rangeland condition score of 52%. The farmer was usi