The role of climate-smart agriculture in strengthening South Africa’s food system and climate response

Climate-smart agriculture (CSA) has become a prominent framework for addressing the interconnected challenges of climate change, food insecurity and environmental degradation. In its conventional formulation, CSA seeks to advance three broad objectives: sustainably increasing agricultural productivity and incomes, strengthening adaptation and resilience, and, where feasible, contributing to mitigation. However, these objectives do not automatically align in practice. What counts as “climate-smart” is highly context-specific, and the effects of CSA depend not only on agronomic techniques, but also on institutional support, market access, land relations, knowledge systems and local ecological conditions (FAO, 2017; Karlsson et al., 2018). This question is especially important in South Africa, where agriculture remains highly exposed to recurrent drought, rising temperatures, erratic rainfall and intensifying water stress. Recent work on crop production, rural livelihoods and drought in South Africa underscores the growing pressure of climatic variability on agricultural productivity and food security (Olabanji et al., 2020; Nyahunda and Tirivangasi, 2022; Zenda, 2024; Letswamotse et al., 2024). National adaptation planning similarly identifies agriculture, food security, water and rural livelihoods as closely interlinked climate-vulnerability priorities, underscoring the need for integrated responses that strengthen resilience while sustaining production systems (DFFE, 2020). In this setting, the significance of CSA lies not merely in its technical promise, but in its potential to support context-appropriate responses that improve food-system resilience under conditions of unequal resource access and uneven institutional capacity (DFFE, 2020; FAO, 2017).

For South Africa, this matters because food security cannot be reduced to production alone. A more robust assessment must consider food availability, access, utilisation and stability together. From this perspective, agricultural interventions should be evaluated not only in terms of yield effects, but also in relation to whether they improve household access to food through income and affordability, support dietary quality and nutritional outcomes, and reduce vulnerability to climatic and economic shocks over time. CSA is often presented as well suited to this integrated agenda, yet the empirical basis for such claims remains uneven. Some interventions appear to support productivity and adaptation under specific conditions, whereas others generate more limited, indirect, or highly contingent benefits (FAO, 2017). South African evidence likewise shows that although CSA practices can improve soil health, water management, productivity and resilience, adoption and outcomes remain strongly shaped by farmers' resource endowments, education, institutional support and local agro-ecological conditions (Oduniyi and Sylvia, 2019; Molieleng et al., 2021; Machete et al., 2024).

A further complication is that the CSA literature is not uniformly affirmative. Although many studies report positive associations between CSA practices and productivity, resilience or household welfare, broader review scholarship increasingly emphasises uneven adoption, variable outcomes, and unresolved questions about scalability, equity and evidence strength. Critical work has shown that the language of CSA can obscure trade-offs, distributive tensions and implementation constraints, especially when claims of “triple wins” are advanced without sufficient attention to who benefits, under what conditions and with what consequences (Karlsson et al., 2018). These concerns are particularly relevant in South Africa, where agrarian inequality, extension deficits, insecure land tenure and uneven state capacity are likely to shape both the uptake and the effects of CSA interventions. Existing South African studies already point to the importance of access to credit, extension services, irrigation support, information and institutional coordination in determining whether farmers are able to adopt and sustain CSA practices (Senyolo et al., 2018, 2021; Serote et al., 2021, 2023). The existing review literature nevertheless leaves an important gap. One strand of scholarship provides broad regional overviews of CSA adoption and implementation in sub-Saharan Africa. These reviews are useful for identifying general patterns, but their breadth limits their capacity to offer a detailed country-specific synthesis of South Africa's empirical evidence base (Kombat et al., 2021; Zenda and Malan, 2021, 2024). Another strand focuses more narrowly on adoption, scaling, institutional mechanisms, policy frameworks or market dynamics. In the South African case, Olabanji and Chitakira (2025), for example, review the adoption and scaling of CSA innovations among smallholder farmers, with particular attention to institutional mechanisms, policy frameworks and market dynamics. While this is an important contribution, it is not a systematic empirical synthesis of how CSA practices in South Africa relate simultaneously to food security outcomes, adaptation effects, mitigation co-benefits and barriers to implementation. Consequently, despite a growing body of country-specific studies, there remains a limited synthesis that integrates these dimensions within a single analytical frame (Olabanji and Chitakira, 2025).

That gap is analytically significant for three reasons. First, South African CSA research is dispersed across provinces, farming systems and methodological traditions, making it difficult to assess where evidence is concentrated and where important blind spots remain. Second, many studies emphasise adoption, productivity or resilience, but fewer explicitly connect CSA interventions to the four dimensions of food security. Although some South African studies have established positive relationships between CSA adoption and household food security, the evidence remains fragmented across practices, regions and outcome measures (Oduniyi et al., 2022). Third, adaptation outcomes are more commonly reported than mitigation outcomes, while direct greenhouse-gas measurement remains much less common than inference from soil management, reduced input use or ecosystem restoration. Without a systematic synthesis that distinguishes robust findings from more inferential claims, there is a risk of overstating the empirical basis of CSA while underestimating the institutional conditions required for effective implementation (FAO, 2017). This study addresses that gap through a systematic review of recent empirical evidence on CSA in South Africa. Rather than offering a broad regional synthesis or an adoption-only review, it examines how CSA practices reported in the South African empirical literature relate to food availability, access, utilisation and stability, and how they contribute to climate adaptation, mitigation co-benefits, and broader climate response. It also identifies the institutional, socioeconomic and behavioural factors that shape adoption, including extension support, land tenure, access to finance, gender disparities and youth disengagement. By integrating these dimensions, the study seeks to clarify what the current South African evidence supports, where benefits are most consistently documented, where claims remain more tentative and which constraints most strongly shape implementation outcomes (Olabanji and Chitakira, 2025; DFFE, 2020).

Conceptually, the review is guided by a socio-ecological and governance perspective. This framing is more appropriate to the present study than an expansive list of loosely connected theories, because it treats CSA outcomes as emerging from the interaction between environmental stress, farming systems, institutions and farmer agency. Such a perspective makes it possible to analyse CSA not merely as a bundle of technical practices, but also as an implementation process shaped by policy coherence, advisory systems, market access, local knowledge and unequal access to productive resources. This conceptual stance is therefore well aligned with environmental management scholarship concerned with sustainability, adaptation and the governance of socio-ecological transitions (DFFE, 2020; Hermans et al., 2020). This review addresses three interrelated questions: how these practices relate to the four dimensions of food security; what evidence exists for adaptation outcomes and mitigation co-benefits and which barriers and enabling conditions most strongly influence adoption and implementation? Answering these questions helps position CSA more realistically within South Africa's food-system and climate-policy landscape and provides a stronger basis for evidence-informed policy and practice. From a theoretical perspective, this contributes to environmental management, sustainability and climate adaptation scholarship by reinforcing the importance of multi-level governance and socio-ecological interactions in shaping adaptation outcomes.

The present research conducted a systematic review practice that follows the PRISMA 2020 (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework and principles, which ensure transparency, rigour and reproducibility of evidence and synthesis (Page et al., 2021; Brennan and Munn, 2021). The procedure of review included several steps: the eligibility criteria were established, an extensive literature search was performed, and the screening/selection, data extraction and synthesis steps were used. A total of 213 records were identified initially using the electronic databases and other sources (Figure 1). The sources were Scopus, Google Scholar, ScienceDirect and institutional repositories, according to the details under broad-based search provided by PRISMA-S (Rethlefsen et al., 2021). The search strategy involved keywords like: climate smart agriculture, climate smart, food security, food systems, climate change, climate variability, CSA practices, adaptation and mitigation or barriers to CSA, South Africa and other Boolean operators to narrow the search. Each database had search strings specific to it and optimised to offer higher specificity and yield. This procedure led to the identification of 30 studies, as illustrated in Figure 1 (PRISMA Flow Diagram), that satisfied the eligibility criteria.

In this review, systematic reviews were retained as supplementary sources of evidence to provide contextual understanding and to complement the primary empirical findings, rather than to serve as primary units of analysis. To avoid double-counting and circular reasoning, data extraction and synthesis were restricted to primary empirical studies only, while findings from included systematic reviews were used for contextual comparison and triangulation of themes. This approach ensured that no overlapping datasets were directly analysed more than once, thereby maintaining the integrity and validity of the synthesis.

Inclusion criteria required that studies:

1. Dwelling upon CSA practices in the South African setting,

2. Tackled either of the two problems, food security and climate change,

3. Were published in English between 2018 and 2025, and

4. Used empirical or systematic review methods.

Exclusion criteria ruled out studies that:

1. Did not include South Africa as a focal region,

2. Were non-empirical (e.g. opinion pieces, editorial notes),

3. Did not address CSA interventions or their outcomes.

A standardised form informed data extraction by including information such as publication year, CSA practices, methodology, geographic focus, type of study, population/sample as well as food security or climate-related outcomes. All the studies were grouped into three types of methodology: quantitative (50%), mixed methods (30%) and qualitative (20%). In the qualitative studies, thematic analysis was conducted using Braun and Clarke's six-phase reflexive approach, with the phases treated as recursive rather than strictly linear (Braun and Clarke, 2021). This process allowed identifying the common themes on CSA implementation, barriers, enablers and impact. An inductive coding technique was used to adjust to the new knowledge, and deductive coding was used to adhere to the existing CSA and food security frameworks. Descriptive mapping of the frequency and distribution of methodologies, crop-livestock systems and geographic coverage was synthesised based on quantitative data. An integrated synthesis of mixed-methods research was conducted with a view to thorough interpretation of the results.

Table 1 presents the main search framework. The search strings were implemented using a date range of 2018–2025, tailored to match the specific syntax requirements of each database, and restricted to English-language results. The selection of the review period from 2018 to 2025 is grounded in the need to capture the most recent and policy-relevant developments in CSA, particularly within the rapidly evolving context of climate change and food systems in South Africa. This period aligns with the growing global and national emphasis on climate resilience following the adoption and early implementation phases of the Paris Agreement and the Sustainable Development Goals, which significantly accelerated research, funding and policy interventions related to CSA.

The quality appraisal and risk-of-bias assessment were conducted independently by two reviewers using established critical appraisal tools appropriate to the respective study designs (Table 2). This independent assessment process was implemented to enhance consistency, reduce subjective bias and strengthen the reliability of the review findings. Any discrepancies identified during the appraisal process were resolved through discussion, and where consensus could not be achieved, a third reviewer adjudicated the final decision. This approach ensured methodological rigour, transparency and alignment with best practices for systematic reviews under the PRISMA 2020 framework. Each of the 30 included studies was systematically evaluated using established critical appraisal tools aligned with study design. Quantitative studies were assessed for internal validity, sampling adequacy and measurement reliability; qualitative studies were appraised based on criteria such as credibility, transferability, dependability and confirmability; while mixed-methods studies were evaluated for the integration and coherence of qualitative and quantitative components. Quality appraisal was conducted independently by two reviewers using design-appropriate critical appraisal tools, including the Joanna Briggs Institute (JBI) critical appraisal checklists, the Critical Appraisal Skills Programme (CASP) tools, Cochrane risk-of-bias assessments, and the Mixed Methods Appraisal Tool (MMAT) for mixed-methods studies. Disagreements were resolved through discussion, and where consensus could not be reached, a third reviewer adjudicated. The appraisal outcomes were not used to exclude studies automatically but informed the interpretation of the evidence, with greater analytical confidence placed in findings from studies assessed as methodologically stronger.

Each study was assigned a quality rating (e.g. high, moderate or low), and potential sources of bias, such as selection bias, reporting bias and methodological limitations, were identified and documented. The results of this appraisal were used to guide the interpretation and synthesis of evidence, ensuring that greater weight was given to methodologically robust studies. This additional step enhances the transparency, rigour and reproducibility of the review, aligning more fully with the expectations of PRISMA and best practices in systematic review methodology.

This systematic review has chosen 30 studies, which indicates an increasing academic focus on the use of CSA to address food security and climate change problems in South Africa. Temporal distribution of the studies reveals that the number of studies is growing over the years with most (10 studies) published in 2024 (Figure 2). This is preceded by 7 studies in the year 2021 and 4 in 2022, 3 each in 2020 and 2025, 2 in 2018, whereas there was only one study in the year 2023. As Figure 2 indicates, this trend implies an increased focus on CSA over the past few years, which is possibly the reaction to the growing climate-related agricultural vulnerabilities and the necessity of more sustainable agricultural practices.

When looking at methodological practices, the review shows that quantitative methods were by far the most widely used, making up about half (50%) of all studies (Figure 3). This suggests that many researchers leaned towards approaches that emphasise numbers, statistics and measurable outcomes. Quantitative methods are often chosen because they allow researchers to work with larger samples, apply objective analysis and generate findings that can be generalised more broadly. In most cases, these studies relied on tools such as surveys, structured questionnaires, econometric modelling and statistical analysis to test hypotheses, identify patterns or explore correlations.

Mixed methods approach made up 27% of the studies, reflecting a considerable share of research that seeks to bridge the gap between numbers and narratives. The adoption of mixed methods highlights an awareness among scholars that some questions cannot be fully understood through numbers alone. By combining quantitative techniques with qualitative ones, such as interviews or focus group discussions, these studies captured both broad statistical trends and the deeper social or cultural factors shaping them. This approach provided a more holistic perspective on complex issues. Qualitative methods accounted for 23% of the studies. While smaller in number, these studies played an important role in offering detailed and nuanced insights that go beyond what numerical data can reveal. Using methods such as in-depth interviews, case studies, participant observation and discourse analysis, qualitative research was particularly valuable for unpacking context-specific dynamics, understanding lived experiences and uncovering local knowledge systems. Instead of focusing on measurement, these studies often aimed to generate fresh insights or theories that enrich the broader research field.

The majority of the research was empirical, and data were based on household survey, structured questionnaire, focus group, participatory learning and field experiment. The size of samples was wide ranging, with some focused-on samples of a community level to studies that were national in scope or extended to many provinces and a sample size of thousands of smallholders. Two studies only relied on review approaches by using literature synthesis and policy analysis and their research orientation was at the national or regional agricultural systems. Figure 4 highlights that the studies covered a range of climatic zones, including semi-arid regions, sub-tropical areas, and, in some cases, even warmer regions. From a geographic perspective, the distribution of research was fairly balanced across South Africa's six major provinces, though certain regions received more attention than others.

The highest concentration of studies was found in Limpopo, with six studies, reflecting the province's importance as a predominantly rural and agriculturally dependent area where CSA practices are particularly relevant. Mpumalanga and studies conducted at the national level followed closely with five studies each, showing strong interest both in region-specific and countrywide perspectives on CSA. In the North West and KwaZulu-Natal, there were three studies each, suggesting moderate research engagement in these provinces, likely due to their mix of commercial and smallholder farming systems. The Eastern Cape had two studies, which align with the province's diverse farming conditions but may also point to gaps in CSA-focused research there. Finally, the Free State and Gauteng recorded the least attention, with one study each, highlighting significant underrepresentation in these areas. This provincial distribution underscores how CSA research has been conducted in varied agro-ecological zones, reflecting the diversity of farming systems, climate challenges and adaptation needs across South Africa. It also points to areas where research is relatively limited, suggesting opportunities for further study, especially in provinces with fewer documented CSA initiatives.

A key strength of this study lies in its inclusion of a diverse and heterogeneous body of evidence on CSA across multiple agro-ecological contexts in South Africa. By synthesising studies with varying designs, sample sizes and analytical approaches from small-scale qualitative case studies to large-scale surveys and advanced econometric analyses the review captures a broad and comprehensive understanding of CSA adoption and effectiveness. This diversity allows for richer insights into context-specific dynamics and enhances the overall depth of analysis, enabling more nuanced and inclusive conclusions that reflect the complexity of farming systems and climate adaptation challenges. Figure 5 illustrates the proportional distribution of crops and livestock types in selected CSA studies. In the agricultural systems under study, 46% of the work focused on mixed crop-livestock systems, highlighting the growing relevance of integrated practices in the implementation of CSA. The crop-only practices were included in the 47% of studies as opposed to livestock-based practices that were just 7%. As indicated in Figure 5, the majority of systems are mixed systems, which underpins the emphasis of CSA on resiliency and eco-balance by diversifying. Maize, legumes, vegetables, sunflower and livestock systems such as cattle, goats and poultry, were by far the most common crops and livestock systems studied, respectively. This variety reflects the adaptability of CSA to both subsistence and semi-commercial farming contexts, reinforcing its value in enhancing productivity, resilience and sustainability in vulnerable agricultural communities.

“Typology” refers to a structured way of classifying elements according to their common features as illustrated in Table 3. It helps in identifying distinct patterns or components within a given field, making it easier to understand complex issues. By organising knowledge in this way, typologies support meaningful comparisons and provide a useful guide for research and analysis across different disciplines, including sociology and agriculture (Mizik, 2021; Thi and Zhou, 2025; Tirivangasi, 2024). In CSA, typology is an important tool for grouping farming practices that promote resilience and sustainability in the face of climate change. These practices are generally organised into three main categories: adaptation, mitigation and productivity enhancement. Each category includes a range of techniques that offer tailored solutions to the challenges brought about by climate variability (Mwongera et al., 2017). These categories will be explained below.

(1) Adaptation Practices: These focus on strengthening farmers' ability to cope with the impacts of climate change. Examples include using drought-tolerant crop varieties, harvesting and storing rainwater, and adjusting planting times. Such measures reduce vulnerability, safeguard food security and help communities adapt to shifting climate conditions (Murray et al., 2016; Mwongera et al., 2017; Phiri et al., 2022). (2) Mitigation Practices: This category emphasises reducing greenhouse gas emissions from farming activities (Staden, 2020). Techniques such as conservation tillage, agroforestry and better manure management not only cut emissions but also improve soil health and increase carbon storage. In this way, mitigation strategies contribute to both climate protection and long-term agricultural sustainability (Arenas-Calle et al., 2019; Staden, 2020). (3) Productivity Enhancement Practices: These practices are designed to increase yields in a sustainable manner. Methods like crop rotation, integrated pest management and the use of organic fertilisers boost efficiency, maintain biodiversity and improve soil fertility. As a result, they raise productivity without compromising environmental health (Tiamiyu et al., 2018; Aryal et al., 2018). CSA practices in South Africa are heterogeneous, context-dependent and shaped by both socioeconomic conditions and environmental factors. In Latin America, differences in the uptake of CSA practices are closely associated with farmers' perceptions of climate change and their assessment of the effectiveness of CSA interventions, both of which are shaped by local agricultural policies and community-based initiatives (Acosta et al., 2021). Farm-level adoption of CSA is a prevailing theme across literature, often motivated by local knowledge, climate risk and institutional backing like extension services. A key set of CSA practices includes crop diversification and better cultivars. These encompass rotations, intercropping and planting drought-tolerant or early maturing varieties. Research by Obi and Maya (2021), Makamane et al. (2023), and Machete et al. (2024) shows that these measures boost productivity and resilience in semi-arid and drought-prone regions. The practices hold particular significance in regions such as the Eastern Cape and Limpopo, where climatic stress is increasing and farmers use diversification to cushion against shocks.

Organic manure use also became popular as an affordable, available practice to replenish soil fertility and cut down on chemical inputs. Based on the work of Chitakira and Ngcobo (2021), farmers perceived organic manure to be not only technically feasible but also socially and economically acceptable. It was most common in KwaZulu-Natal and the Eastern Cape, where manure is easily available from mixed farming systems. Water management practices, especially small-scale irrigation and rainwater harvesting, were prominent adaptive reactions in water-scarce areas like Gauteng and Mangaung. Chitakira and Ngcobo (2021), and Makamane et al. (2023), Kubanza and Oladele (2024) highlighted the increased use of drip and sprinkler irrigation to compensate for unpredictable rainfall and lengthen the growing season. Not only do these technologies increase productivity, but they also help with adaptation. CSA also includes crop-livestock integrated systems that recycle nutrients and diversify incomes. These were observed in small-scale and commercial farms in the Eastern Cape, Limpopo, and Mediterranean regions (Obi and Maya, 2021; Swanepoel and Smit, 2025). Farmers who adopted the systems reported enhanced soil health, reduced application of synthetic inputs and higher efficiency of pasture land utilisation. Agroforestry, often founded on local knowledge, integrates ecological restoration and agricultural productivity. It was widely adopted in Limpopo, Eastern Cape and national programs (Senyolo et al., 2018; Ayisi et al., 2021; Zerihun, 2021). Farmers integrated species like Moringa oleifera and Vachellia Karoo into crop systems to provide fodder, improve soil fertility, and cushion against weather variability. Several studies pointed to soil and water conservation techniques, including mulching, cover cropping and crop rotation, which were widely used to minimise erosion, increase water retention and boost fertility (Obi and Maya, 2021; Ndlovu et al., 2022). These are especially applicable in semi-arid and erosion-prone regions such as Bushbuckridge and KCDM. Lehmann et al. (2020) further highlight that agroforestry systems in Italy and Denmark enhance farm productivity while simultaneously supporting biodiversity conservation by deliberately combining trees with crop production systems.

Drought-tolerant seed varieties (DTSVs), including hybrid, open-pollinated and conventionally bred drought-adapted seeds, were identified as an important CSA practice aimed at enhancing productivity and climate adaptation. Evidence from Limpopo indicates that the adoption of DTSVs contributes to improved crop performance under water-stressed conditions, thereby supporting farmers' adaptive capacity in the face of increasing climatic variability (Senyolo et al., 2021). These findings highlight the role of improved seed systems in strengthening agricultural resilience, particularly in drought-prone environments. Water-smart agriculture (WaSA) focuses on improving water use efficiency through the adoption of basic water conservation techniques aimed at enhancing climate adaptation. In Bushbuckridge, evidence indicates that WaSA practices contribute to improved water management and increased resilience to water scarcity among smallholder farmers (Ndlovu et al., 2022). These findings suggest that even low-cost, context-specific water-saving interventions can play a significant role in strengthening adaptive capacity in drought-prone rural farming systems.

Regenerative agriculture is an ecosystem-based farming approach that reduces chemical inputs while enhancing soil health, biodiversity and overall system functionality. Evidence from degraded agricultural zones at the national level indicates that regenerative practices contribute to the three pillars of CSA productivity, adaptation and mitigation by restoring soil fertility, improving ecosystem resilience and supporting more sustainable production systems (Moodley et al., 2024). These findings highlight the potential of regenerative agriculture as a holistic strategy for addressing multiple climate and food security challenges simultaneously. Agri-silviculture, which involves the integration of trees and crops on communal land, was identified as a CSA practice that supports both adaptation and productivity objectives. Evidence from Mpumalanga suggests that this system contributes to improved food production and diversified income streams for farming households, while also enhancing environmental resilience through improved soil conservation and microclimate regulation (Maponya, 2024). These findings underscore the role of tree–crop integration in strengthening the sustainability and resilience of smallholder farming systems.

Conservation agriculture (CA), which consists of minimum tillage, permanent soil cover, and diversified crop rotations, has been at the centre of CSA implementation in areas like Bergville, Mpumalanga and the Western Cape. In regions of Nepal prone to soil erosion and water scarcity, the strategic implementation of these conservation methods ensures that agricultural land remains productive, thereby supporting household food security (Mishra et al., 2020). Kruger et al. (2022) and Oduniyi et al. (2022) illustrated that CA practices not only improved productivity and resilience but also enabled social innovation through institutions like Village Savings and Loan Associations (VSLAs), which enabled farmers to reinvest in sustainable practices. Finally, farmer training, awareness-raising initiatives, and institutional support mechanisms such as extension services and learning groups were identified as important enabling factors that may support the adoption of CSA practices. Senyolo et al. (2021) and Kruger et al. (2022) reveal how the use of participatory tools and sharing of knowledge enhanced adoption, particularly in marginalised or resource-poor communities.

CSA awareness and information access are key enablers of CSA adoption in South Africa. Senyolo et al. (2021), Kruger et al. (2022), and Machete et al. (2024) show that training, extension services, information sharing platforms and participatory learning groups improve farmers' understanding and uptake of CSA practices, especially in Limpopo and KwaZulu-Natal. Nationally, timely and context-specific climate information supports all CSA pillars. However, unequal access to extension services and weak institutional capacity still limit widespread adoption, highlighting the need for more inclusive and decentralised knowledge systems.

Limitations, trade-offs and implementation challenges associated with CSA in South Africa are widely reported across empirical studies and review syntheses, highlighting a consistent pattern of context-specific barriers that constrain broad adoption and effective implementation. A key issue repeatedly identified is the significant information and financial constraints faced by smallholder and communal farmers, who constitute a large share of the agricultural sector. Limited access to extension services, advisory support and timely CSA-related information reduces farmers' capacity to practically implement CSA innovations (Senyolo et al., 2018). These informational challenges are closely linked to financial barriers, including high initial investment costs, restricted access to credit and weak risk-sharing mechanisms, all of which diminish the economic feasibility and attractiveness of CSA for resource-constrained farmers (Kapari et al., 2023; Mutengwa et al., 2023). In addition, variability at the farm and landscape level, such as small landholdings, erratic rainfall patterns and diverse risk conditions, introduces high transaction costs and limits the suitability of standardised CSA interventions, thereby reducing both adoption rates and the potential benefits realised by farmers (Mizik, 2021; Barasa et al., 2021).

CSA interventions have shown substantial and multidimensional effects on food security through improved food availability, access, utilisation and stability (Table 4). In many studies, CSA practices like better seed varieties, crop diversification, soil and water conservation, and integrated crop-livestock systems have led to improved productivity, resilience and household food security. Food Availability was enhanced consistently through CSA's contribution to crop productivity and yield increases. Maize yields, for instance, were almost double CSA adopters compared to non-adopters, according to Omotoso et al. (2024), while CA stabilised yields even during climate stress (Kruger et al., 2022). Soil fertility was increased and production systems diversified through CSA practices such as crop rotation, intercropping, and agroforestry, guaranteeing a more stable food supply. Crop rotation and the use of drought-tolerant seeds ensured that yields were maintained under water-scarce conditions in Gauteng (Chitakira and Ngcobo, 2021), while regenerative agriculture enhanced not only yield quantity but also crop nutrient density (Moodley et al., 2024).

Food Access was improved through the increased incomes attained by CSA adoption. Income disparities and increased maize productivity among adopters and non-adopters, R7834.91 against R4904.35, attest to this effect (Omotoso et al., 2024). The income impacts of CSA were also proven by Oduniyi et al. (2022), where minimum tillage and crop diversification resulted in a 60.31% increase in income. Additional evidence provided by Maponya (2024) shows that agri-silviculture increased household income by up to 42%, which had a direct impact by translating to increased purchasing power for food. Access was also increased through the incorporation of VSLA (Kruger et al., 2022) and off-farm sources of income, both of which enabled the increased affordability of food. Food utilisation, which includes nutritional status and the quality of the diet, also reaped benefits from CSA. HDDS scores, a measure of dietary diversity, were found to be significantly higher for adopters, with Omotoso and Omotayo (2024) noting a 27% increase in HDDS compared to non-adopters. Likewise, livestock integration within CSA systems (Ayisi et al., 2021; Swanepoel and Smit, 2025) enhanced access to high-quality protein sources and improved animal health, indirectly contributing to better household nutrition. Organic inputs and soil fertility practices such as compost and biochar (Senyolo et al., 2018; Moodley et al., 2024) increased the nutrient density of food, tackling undernutrition in marginal regions.

Food Stability was supported by CSA's buffering potential against environmental shocks. Practices like WaSA (Ndlovu et al., 2022) and drought-resistant seed varieties (Senyolo et al., 2021) guaranteed steady food stocks even amidst erratic rainfall. Households practicing CSA had lower Household Food Insecurity Access Scale (HFIAS) scores, which reflected reduced food insecurity volatility (Omotoso et al., 2024). Besides, the long-run usefulness of CSA was realised through outcomes such as stable provisioning between 1 and 3 months to 6–12 months following implementation (Kruger et al., 2022), which reflected household resilience. Training and availability of information were similarly effective in triggering adoption and, by extension, stability (Machete et al., 2024). Still, constraints of low awareness, low uptake of some practices (e.g. wetland utilisation, CA), and adverse impacts of large household size continue to limit more extensive CSA impact. These results highlight the need for localised training, farmer extension services and the promotion of context-specific practices to completely unleash the food security potential of CSA.

CSA practices were found to contribute substantially to climate change adaptation across South Africa's diverse agro-ecological regions, while evidence regarding their mitigation impacts was more limited, indirect and less consistently quantified (Table 5). While few papers presented quantified greenhouse gas (GHG) emission data, numerous contributions demonstrated indirect or inferred mitigation through improved soil health, reduced synthetic inputs and improved carbon sequestration (Branca et al., 2021; Moodley et al., 2024). Adaptation benefits were more widely documented, with strong evidence of increased resilience to drought, improved productivity, and enhanced food security for smallholder farmers. One key contribution is in soil and water-based practices. Crop rotation, organic manure, mulching, and cover cropping were commonly taken up to enhance soil fertility, water retention and erosion reduction. These were observed to build resilience in semi-arid and peri-urban environments. Kruger et al. (2022) determined that CA reduced tillage, crop rotation and permanent soil cover led to significant organic carbon (SOC) gains (up to 24%) and drought tolerance improvement. Abegunde et al. (2022) also noted that organic manure and crop-livestock integration sustained soil health and stability, especially in KwaZulu-Natal's mixed farming systems.

A review focusing on Southeast Asia highlighted the ability of CSA to enhance resilience through the implementation of innovative techniques like regenerative soil management that boosts organic carbon levels and improves water retention in soils (Udomkun et al., 2025). Such practices are essential in regions experiencing erratic rainfall and extreme weather conditions, thereby supporting farmers' adaptation strategies to climatic changes. CSA also contributes through biodiversity-enhancing practices. Agroforestry emerged as a key strategy, especially in the Eastern Cape and Limpopo. Studies by Ayisi et al. (2021) and Zerihun (2021) revealed that integrating trees into farming systems enhanced biodiversity, provided ecosystem services and supported GHG mitigation through carbon sequestration and methane reduction. Tannin-rich fodder trees, such as Vachellia karroo, were especially effective in lowering enteric emissions from livestock. Farmers' behavioural responses and knowledge systems were at the core of CSA's success in climate adaptation. CA and DTSVs were some of the practices that allowed farmers to act in response to increasingly unpredictable rainfall. Senyolo et al. (2021) and Machete et al. (2024) emphasised how climate awareness and access to CSA information impacted the uptake of adaptive strategies. Training, extension services, and participatory innovation platforms were crucial in driving these outcomes. Crop-livestock integration and regenerative agriculture were further contributions with adaptation and mitigation impacts. Swanepoel and Smit (2025) highlighted the role of nutrient recycling in mixed systems in decreasing emissions and enhancing soil structure. Moodley et al. (2024) showed that regenerative methods restored degraded soil and developed drought tolerance, particularly in dry areas. These strategies, while not necessarily formally quantified, were consistent with climate objectives by decreasing chemical inputs, enhancing productivity and facilitating ecosystem restoration.

Water-smart practices also built resilience in drought-prone regions. Rainwater harvesting, drip irrigation, and water-use efficiency technologies like those researched by Ndlovu et al. (2022) and Kubanza and Oladele (2024) were key factors in semi-arid regions like Bushbuckridge and NMMDM. Although GHG reductions were not measured, these practices stabilised farm outputs during water stress conditions. In total, South African CSA practices have also functioned as successful adaptation strategies, with secondary mitigation co-benefits in the form of enhanced soil carbon and reduced input dependence. Although quantifications of GHG savings remain few, the practices' faithful adherence to CSA's three pillars of productivity, resilience and mitigation indicates strong potential for broader climate-smart development.

Limitations to the uptake of CSA emerged in a number of thematic domains, from institutional, socioeconomic, to behavioural challenges (Table 6). Among the most frequent institutional limitations were inefficient extension services, mentioned in multiple studies. Farmers frequently reported insufficient access to trained extension agents, delayed delivery of inputs and no follow-up support (Obi and Maya, 2021; Senyolo et al., 2021; Omotoso et al., 2024). This was supplemented by fragmented or poorly harmonised policies, with farmers being either unaware of CSA policies or receiving confusing information due to compartmentalised government activities (Zerihun, 2021; Kubanza and Oladele, 2024).

Policy gaps and weak coordination significantly hinder the effectiveness of CSA implementation, with Kubanza and Oladele (2024), Zerihun (2021), and Senyolo et al. (2018) noting disjointed policies, limited interdepartmental coherence and unclear implementation guidelines. These challenges often result in fragmented programmes and inconsistent support for farmers, reducing the overall impact of CSA initiatives. To improve coherence, there is a need for bottom-up policy formulation processes, consolidated and well-aligned programmes, and the development of localised CSA guidebooks that reflect contextual realities. Strengthening policy integration would also benefit from improved CSA valuation and cost–benefit analyses to support evidence-based decision-making and more efficient allocation of resources.

Insecurity of land tenure particularly for women constituted another structural hurdle. In Limpopo and NMMDM research, insecure landholding dissuaded long-term investment in CSA technologies (Ayisi et al., 2021; Senyolo et al., 2021). Furthermore, insufficient training, specifically for agro-ecological or regenerative approaches, was a prevalent grievance. Numerous farmers, particularly older farmers, perceived CSA approaches as complicated or in conflict with legacy systems upon which they had depended for many years (Moodley et al., 2024; Swanepoel and Smit, 2025).

Socioeconomic limitations also constrained adoption. Low income, high labour demands, and small farm sizes disproportionately affected women-headed and resource-poor households (Kruger et al., 2022; Msweli et al., 2024). Education was a key discriminating variable: farmers with more years of education were much more likely to adopt practices such as crop rotation and integrated livestock systems (Abegunde et al., 2022). Cultural mismatch was also present, wetland utilisation, for instance, was rejected in some environments despite environmental benefits, due to social taboos or technical complexity (Abegunde et al., 2020). Behavioural and psychological barriers also influenced CSA uptake. Farmers' perceived riskiness, low self-efficacy and preference for short-term gains over long-term resilience discouraged adoption (Senyolo et al., 2018; Omotoso and Omotayo, 2024). Ultimately, attitudes, norms and peer pressure influenced uptake, suggesting that social learning and group-based training can overcome some barriers.

Labour and input constraints present significant barriers to CSA adoption, as highlighted by Makamane et al. (2023), Moodley et al. (2024), and Oduniyi et al. (2022), who note challenges such as labour shortages, rising input costs, soil degradation and declining soil organic matter. These factors reduce productivity and limit farmers' ability to implement labour-intensive or input-dependent practices. Addressing these constraints requires promoting regenerative agricultural practices, providing CSA training tailored to low-labour systems and enhancing soil health through organic amendments and other soil restoration techniques. Future work should also focus on monitoring ecosystem services and evaluating the effectiveness of regenerative agriculture interventions in improving long-term soil fertility and farm resilience.

Behavioural resistance and risk perception significantly influence CSA adoption, with Omotoso and Omotayo (2024) highlighting farmers' perceptions of CSA as complex, low confidence in new practices and a preference for familiar traditional farming systems. These psychological and behavioural factors often slow down the uptake of innovative practices even when their benefits are well demonstrated. To overcome this, interventions such as peer learning networks, behavioural nudges, farmer field schools and on-farm CSA demonstrations have been recommended to build trust, improve confidence and enhance experiential learning.

Disengagement of youth and gendered disparities is a challenge in semi-arid regions. Young people often found farming unappealing due to low status and returns, while women, who were highly involved in farming, lacked institutional support, access to credit and training (Ndlovu et al., 2022; Maponya, 2024). Several studies recommended targeted interventions, youth-specific training, gender-sensitive extension services and land rights reform to make CSA more inclusive and appealing. With these limitations in mind, most authors called for context-tailored policies, strengthened extension services, CSA-tailored financing mechanisms and robust impact evaluation tools. Both long-term research and behavioural research were also pinpointed as critical research gaps for tracking adoption patterns over time and sharpening interventions as we move forward.

Cultural and technical misalignment poses a significant barrier to CSA adoption, as Abegunde et al. (2020) and Zerihun (2021) note that certain CSA practices may conflict with local cultural beliefs and practices, such as restrictions on the use of wetlands, while others face practical implementation challenges, including agroforestry systems. These mismatches reduce farmer acceptance and limit the effectiveness of otherwise beneficial interventions. To address this, there is a need for culturally sensitive CSA design that aligns with local norms and prioritises practices with higher social acceptance. Strengthening understanding of locally acceptable technologies, particularly in agroforestry, can improve uptake and sustainability.

This systematic review demonstrates that CSA can strengthen South Africa's food system while supporting adaptation to climate change. The evidence indicates that crop diversification, CA, organic soil management, agroforestry, water-efficient practices and integrated crop–livestock systems can improve productivity, soil health, household income, dietary diversity and resilience to climatic shocks. However, adaptation benefits are more consistently documented than mitigation outcomes, which are largely inferred from improved soil carbon, nutrient recycling, ecosystem restoration and reduced dependence on synthetic inputs. The findings reinforce a socio-ecological and governance perspective by showing that CSA outcomes depend not only on agricultural technologies but also on interactions among environmental conditions, farmer agency, institutional support, knowledge systems, land relations and access to productive resources. This interpretation is consistent with insights from agro-ecological, sustainable-livelihoods and socio-technical-transition scholarship (Sachet et al., 2021; Durán et al., 2023). Scaling CSA therefore requires coordinated and context-sensitive interventions. Agricultural extension services should be decentralised and strengthened through demonstration plots, participatory learning, peer exchange, digital advisory systems and locally relevant climate information. Policies should support integrated farming systems and expand access to finance, insurance, inputs, markets and technical assistance, particularly for smallholder farmers. CSA should also be incorporated into national and provincial agricultural, land reform, food security, rural development and climate adaptation strategies. Future research should employ longitudinal and robust impact-evaluation designs, develop standardised indicators and directly measure mitigation outcomes. Greater attention should be given to underrepresented provinces and to the experiences of women, young people and marginalised farmers. Investment in water infrastructure, soil conservation, watershed management and ecosystem restoration will be essential for achieving equitable and sustainable CSA implementation.