University of Eswatini

Illegal trade in leopard tortoises Stigmochelys pardalis is widespread, yet the origins of confiscated S. pardalis are often unknown. Accurate assignment to source populations is essential for informed conservation management and release. We expanded and tested a genetic reference database for assigning S. pardalis to their original wild populations in KwaZulu-Natal Province, South Africa, and Eswatini. We added 50 newly genotyped wild S. pardalis sampled across these regions to the database and then evaluated the updated database by sampling and assessing 50 S. pardalis of unknown geographic origin held at a rehabilitation centre in KwaZulu-Natal. Genetic assignment indicated that only a single S. pardalis at the rehabilitation-centre may have originated from KwaZulu-Natal, highlighting the extent of illegal trade, translocation, and possession of S. pardalis. Individuals of verified regional origin were subsequently rehabilitated and translocated for soft release at appropriate sites.

Municipal solid waste (MSW) management remains a persistent sustainability chal-lenge in low- and middle-income countries, where uneven service coverage and rapid spatial change produce heterogeneous household disposal behaviours and substantial environmental externalities. This study develops a spatially explicit Bayesian network framework to map and explain six dominant household solid-waste disposal pathways across Eswatini using enumeration areas (EAs; n = 2,326) and nationally consistent census-linked predictors. Separate Tree-Augmented Naïve Bayes (TAN) models were trained for regular collection, irregular collection, open burning, public dumping, backyard pit disposal, and undesignated disposal, integrating socio-demographic, in-frastructural, accessibility, environmental, and neighbourhood-context variables, while explicitly quantifying predictive uncertainty using posterior entropy and Kull-back–Leibler (KL) divergence. Hold-out evaluation (465 test EAs; 1,861 training EAs) shows strong pathway-specific performance, with overall accuracy ranging from 0.497-0.989 across targets and ex-pected-value prediction errors of RMSE = 0.148-0.289 and MAE = 0.141-0.242. Uncer-tainty surfaces reveal low entropy in structurally homogeneous, well-served urban cores and elevated uncertainty in peri-urban transition zones where disposal behav-iours are mixed and services are unreliable. KL divergence highlights a limited subset of EAs where local conditions strongly update national expectations—priority loca-tions for targeted interventions and improved data collection. The framework provides policy-ready, uncertainty-aware evidence to support area-based service planning and sub-national monitoring relevant to SDG 11.6.1 in data-constrained contexts.

Sustainable agri-food system transformation in Africa is critical for addressing the persistent challenges of hunger and malnutrition while promoting sustainability. Here, we investigate how green growth initiatives intersect with the agri-food system transformation in the African Continental Free Trade Area context. We combine World Development Indicators, International Financial Statistics, and OECD data, and we use the Prais-Winsten regression model to analyze the relationship between green growth initiatives and Sustainable Development Goal 2- Zero Hunger indicators across 46 African countries. We show that green growth initiatives positively impact agricultural productivity and cereal yield but negatively affect total fisheries production. Low-income countries, including Malawi, Uganda, Mozambique, and Sierra Leone experience mixed effects, while upper-middle-income countries such as South Africa, Botswana, Namibia, and Mauritius face more challenges. Economic growth, renewable energy, and financial development significantly influence agri-food system sustainability, but challenges like carbon dioxide emissions and deforestation hinder progress. Policymakers should prioritize pro-poor green technology investments, enhance finance accessibility, and implement environmental mitigation measures while fostering trade openness and regional cooperation. In Africa, green growth initiatives enhance the value added per worker in agriculture, forestry, and fisheries, as well as cereal yield, but they have a negative impact on total fisheries production, according to an analysis that combines agricultural and economic data with a statistical approach.

Background of study: Lack of physical exercise has significant influence on the rise of cases of non-communicable diseases (NCDs) among university students globally. University students` engagement in physical excercises tend to decline when compared to previous school years. The decline is also observed in Eswatini`s university students. This study, therefore, described the facilitators and barriers of engaging in physical exercises by tertiary students of one Campus of the University of Eswatini. Methods: A descriptive qualitative research design was employed to describe the facilitators and barriers of physical exercise of ten purposively sampled students. Data was collected through audio-recorded face-to-face interviews utilising a researcher developed interview guide. Data were analysed using thematic analysis of Colaizzi’s descriptive method of data analysis. Result: Four themes emerged from the study findings; 1) Students understanding of physical exercise, 2) Facilitators of physical exercise, 3) Barriers of physical excercise, 4) Strategies to improve physical exercise engagement. Conclusion: Students` physical exercise participation is characterised by facilitators and barriers. Therefore, students need practical support from the university in terms of provision of enough variety of training facilities and equipment within the campus, and time alocated for physical activities within the course of their learning.

In an era of polycrisis – characterized by interconnected environmental, economic, social, and health crises amplified by neoliberal policies – higher education must evolve to equip undergraduates with essential skills beyond traditional technical knowledge. This study, situated at the University of Eswatini, (UNESWA), aimed to identify the essential skills undergraduates require to navigate and contribute positively in such contexts, addressing gaps in curricula that prioritize qualification while neglecting socialization and subjectification, as per Biesta's (2015) educational domains. The purpose was to identify perceived essential skills, curricular deficiencies, and integration strategies to foster transformative learning. Employing a mixed-methods sequential design, the research began with a literature review on skills for complexity. Qualitative data were gathered through semi-structured interviews with five purposively selected education stakeholders, exploring perceptions of skills, gaps, and pedagogies. Quantitative data came from surveys of 225 undergraduates randomly selected through stratified sampling (75% response rate) and 18 educators (90% response rate), using a Likert scale to rank skill priorities and identify gaps. Key findings reveal a consensus on top skills: critical thinking and problem-solving (mean 4.6/5), resilience and adaptability (4.5), sustainability awareness and ethical reasoning (4.4), digital literacy and collaboration (4.3), and civic engagement and social justice (4.2). Surveys indicated 72% of undergraduates view curricula as inadequate for polycrisis, with a bias toward qualification over other domains. Qualitative themes highlighted polycrisis as exposing qualification limitations, deficits in ethical initiative and resilience, curricular bottlenecks fostering passivity, and the need for pedagogical shifts like project-based learning. The study concluded that higher education must balance Biesta's (2015) domains to nurture resilient, ethical agents. An empirically-driven framework integrates skills across qualification (e.g., problem-solving), socialization (e.g., collaboration), and subjectification (e.g., ethical initiative), promoting modular curricula, real-world projects, and reflective assessments. Recommendations include prioritizing subjectification through activities building autonomy and ethics; integrating domains via project-based assignments; reforming assessment to reflective portfolios; and conducting multi-site comparative research to adapt the framework regionally.

Abstract Fertiliser input in annual cropping systems can benefit crops and weeds. The effects of organic and inorganic fertilisers on weed species composition were evaluated across two agroecological zones in Eswatini: the Highveld and the Middleveld. Weed field surveys and weed seed bank studies were conducted over two cropping seasons, 2020/2021 and 2021/2022, in fields managed by smallholder farmers practising maize and cattle production. Fertiliser regimes were cattle manure only, cattle manure plus inorganic fertiliser, and inorganic fertiliser only, replicated six times in each region. The weed survey was conducted using three 25-m transects, and the weed seed bank was assessed at sampling depths of 0–10 and 10–20 cm. Results showed that cattle manure only and cattle manure plus inorganic fertiliser regimes had higher species richness and diversity than the inorganic fertiliser only regime. However, weed seed distribution did not differ between sampling depths or fertiliser regimes. Twelve weed species, including bermudagrass, crabgrass, and goosegrass, were key contributors to the differences in species composition between fertiliser regimes and agroecological zones, and were the most prevalent weeds in the seed bank. The study established that fertiliser type influences weed species composition and diversity. Proper cattle manure composting is essential to mitigate weed proliferation and enhance sustainable maize production. These findings demonstrate the need for integrated weed management and fertiliser strategies to improve crop yield and agricultural sustainability for smallholder farmers.

In this paper, using Bregman distance technique, we introduce aN inertial type algorithm with self - adaptive step size for approximating a common element of the set of solutions of pseudomonotone variational inequality problem and the set of common fixed point of a finite family of generic generalized Bregman nonspreading mapping in a real reflexive Banach space. Furthermore, we prove a strong convergence theorem of our algorithm without prior knowledge of the Lipschitz constant of the operator under some mild assumptions. We also give a numerical example to illustrate the performance of our algorithm. Our result generalize and improve many existing results in the literature.

Artificial Intelligence (AI) in higher education worldwide has been greatly revolutionised by technologies like intelligent tutoring systems, learning platforms with adaptivity, and generative AI assistants [1]. In countries like Eswatini, where higher educational organisations face repeated issues like insufficient resources, large student populations, and variable access to technology, AI holds great potential for enhancing student engagement as well as enabling more personalised learning routes [2]. Three main arenas are explored in this research: (1) the impacts on student engagement and academic success caused by tools powered by AI; (2) obstacles and possibilities in adopting AI in circumstances bound by resources; and (3) student perspectives about how AI impacts motivation, belief in oneself, as well as the broader educational experience. Eswatini creates a distinctive case study in which it is characterised by a minimal digital infrastructure (with an internet penetration level of 36%) as well as a strong cultural bias towards conventional face-to-face approaches towards teaching [3]. With a quantitative methodology, this study uses survey responses collected from 100 student respondents representing all study fields and study years, in combination with institutional student records. The survey tools are generative AI tools (e.g., ChatGPT), smart tutoring tools, and adaptive software like Duolingo. Both descriptive statistical and inferential statistical testing were employed in investigating patterns in engagement, performance outcomes, and adoption barriers. Findings indicate 80% of respondents had enhanced motivation and participation with use of AI while there remain key drawbacks including erratic internet connectivity, few apparatuses, and knowledge gaps in digital literacy. Through mitigating such hindrances, universities in Eswatini and similar environments are in a position to leverage AI as a means for sustainable, inclusive, and effective tertiary education. Consequently, this study adds further to the global literature on AI-in-education by extending knowledge beyond developed nations [4].

Purpose: This paper examines the different realisations of Nasal Consonant sequences (NCs) and their syllabification across various morphosyntactic domains in siSwati. Methodology: The study utilises data collected from descriptive and theoretical studies on siSwati linguistics, relying on native speaker intuition and knowledge of the language for verification. It employs Optimality Theory (OT) to account for the intralinguistic variation in the representation of NCs in siSwati. Within this framework, OT provides a principled account of how phonological constraints interact with morphological structure to yield the observed patterns of NC syllabification in siSwati. Findings: Findings indicate that the syllabification of NC sequences is predictable, with /mu/ truncated forms parsing a syllabic segment; otherwise, siSwati maintains homorganicity between the nasal and the following consonant in various derived and non-derived contexts. Unique Contribution to Theory, Practice and Policy: Previous research on NC syllabification in siSwati presupposes the existence of codas emanating from /mu/ truncation. These studies also focused on /mu/ reduction in siSwati alongside other Bantu languages, mapping the degree of similarity and dissimilarities in NC syllabification patterns in the languages under investigation. The current study provides an in-depth investigation of the morphosyntactic and phonological contexts in which /mu/ truncation occurs. The study argues that NC representation is determined by the interaction between phonology and morphology within the grammar. The analysis considers two phonological contexts: a) NCs derived from /mu/ truncation, and b) those in non-derived environments.