A literature search was conducted on 22 March 2025 using the Web of Science and Scopus databases. Searches were performed across titles, abstracts, and author keywords and used a structured Boolean strategy with three keyword blocks. The first block captured maize systems using the terms “maize” OR “corn” OR “Zea mays”. The second block captured GHG-related studies using “greenhouse gas” OR “GHG” OR “emission” OR “flux*” OR “nitrous oxide” OR “N2O” OR “methane” OR “CH4” OR “carbon dioxide” OR “CO2”. The third block restricted retrieval to Sub-Saharan Africa by including regional terms (“sub-Saharan Africa”, “SSA”, and East/West/Central/Southern Africa**) and the names of Sub-Saharan African countries (e.g., Malawi, Kenya, Uganda, Tanzania, Zambia, Zimbabwe, Nigeria, Ghana, Ethiopia, South Africa, among others). The three blocks were combined using AND (maize terms AND GHG terms AND SSA terms), while synonymous terms within each block were combined using OR, to ensure that retrieved records contained maize-related terms and at least one GHG/emissions term alongside a Sub-Saharan Africa location term.

Records retrieved from the database searches were screened and studies were included if they were peer-reviewed, field-based experiments conducted in Sub-Saharan Africa under maize-based cropping systems and reported quantitative soil–atmosphere greenhouse gas fluxes of CO₂, N₂O and/or CH₄. Studies were excluded if they were conducted outside Sub-Saharan Africa, did not involve maize (Zea mays) systems, or did not report quantitative GHG emissions/flux data. We also excluded studies based on laboratory incubations, modelling/simulation, life-cycle assessment, or other indirect estimation approaches without field flux measurements relevant to maize systems. The full inclusion and exclusion criteria are summarised in Table 1.

The study selection process followed PRISMA 2020 reporting guidelines. All records retrieved from Web of Science and Scopus were exported to Mendeley desktop for reference management and duplicate removal prior to screening. The remaining records were screened in two stages: (i) title and abstract screening to exclude clearly irrelevant studies, followed by (ii) full-text assessment to confirm eligibility against the predefined inclusion and exclusion criteria (Table 1). During full-text screening, reasons for exclusion were recorded (e.g., conducted outside SSA, non-maize systems, non-field approaches such as modelling or laboratory incubations, or absence of quantitative GHG flux/emission data). The number of records retained at each stage—identification, screening, eligibility, and inclusion—is summarised in the PRISMA flow diagram (Figure 2).

For each eligible study, data were extracted into a structured spreadsheet using a predefined extraction template. Extracted fields included bibliographic details (authors, year, journal and DOI), study setting (country/region, site description and climate/rainfall regime where reported), experimental design and measurement approach and agronomic management (cropping system, tillage/residue management, crop diversification such as rotations or intercropping, and fertiliser and/or organic amendment inputs including rates and forms). Greenhouse gas outcomes were extracted as reported (CO₂, N₂O and/or CH₄ fluxes and, where provided, cumulative or seasonal emissions), together with the units and reporting period. Bibliometric analyses of the retrieved literature were conducted in RStudio using Biblioshiny (bibliometrix) to summarise publication trends, dominant research themes, keyword co-occurrence networks, and the geographical distribution of studies. Findings were synthesised using systematic mapping and narrative synthesis, grouping results by gas type (CO₂, N₂O, CH₄), management category (e.g., nitrogen inputs, tillage intensity, residue retention, crop diversification), and environmental context (e.g., rainfall seasonality, soil moisture status and temperature where reported). Emission responses were interpreted in relation to management-driven mechanisms (particularly nitrogen availability and residue/organic matter inputs) and site conditions known to influence GHG production and transport, with attention to context dependence and trade-offs between mitigation potential and productivity.

The literature search identified 1,989 records from Web of Science (n = 1,416) and Scopus (n = 573). After merging records in Mendeley and removing 455 duplicates, 1,531 unique records remained for title and abstract screening. At this stage, 1,503 records were excluded because they did not meet the core scope of the review as defined in Table 1. The main reasons for exclusion at title/abstract screening were studies conducted outside Sub-Saharan Africa (n = 687), studies that were not maize-based or did not provide maize plot-level outcomes (n = 391), and studies that did not report quantitative soil–atmosphere GHG fluxes/emissions (n = 369). A small number of records were excluded because they were life-cycle/carbon footprint assessments without field flux measurements (n = 6) or modelling-only studies without field flux measurements/validation (n = 1). The remaining exclusions (n = 49) comprised other ineligible records evident at title/abstract stage, including non-experimental or non-field-based studies, non-eligible document types, and records that did not meet the inclusion criteria in Table 1 upon closer inspection. Following title/abstract screening, 28 full-text reports were sought for retrieval; three reports could not be retrieved (n = 3). The remaining 25 full texts were assessed for eligibility, and four were excluded because the GHG emissions were reported only as aggregated summaries without extractable plot or treatment level values, however, these studies were still considered qualitatively and cited where relevant in the discussion. In total, 21 studies met all eligibility criteria and were included in the narrative synthesis (Figure 2). All included studies were experimental, field-based investigations conducted in SSA that reported quantitative measurements of at least one greenhouse gas (CO₂, N₂O and/or CH₄) from maize-based cropping systems.

The included studies were unevenly distributed across the Sub-Saharan Africa region (Figure 3). Experimental evidence was concentrated in a limited number of countries, with Kenya accounting for the largest proportion of studies, followed by South Africa and Tanzania. Additional studies were identified from Ethiopia, Zimbabwe, Uganda and Ghana. This geographic clustering reflects the presence of long-term experimental sites and established research infrastructure in these countries. Thematic analysis revealed that most studies focussed on the quantification of N₂O emissions in relation to nitrogen management, with comparatively fewer investigations explicitly measuring CO₂ and CH₄ fluxes. Key research themes included fertiliser application rates, soil fertility management, tillage intensity, residue management, and crop diversification practices. Several studies also integrated assessments of soil carbon sequestration, yield performance, and global warming potential, highlighting efforts to link emissions with productivity outcomes. The distribution of studies indicates a strong emphasis on nitrogen-driven emission processes in maize systems, while integrated assessments of multiple greenhouse gases and long-term mitigation outcomes remain less common.

Further analysis of keyword frequency and co-occurrence reveals how research on GHG emissions from maize systems in SSA has been conceptually framed (Figures 4, 5). The dominant research theme centres on N₂O emissions associated with nitrogen inputs, highlighting the central role of nitrogen availability and cycling in regulating climate impacts of maize production. This focus reflects broader concerns regarding fertiliser-driven emission increases under intensification scenarios. A second major theme relates to soil and crop management practices, including tillage intensity, residue retention, crop rotation, and intercropping. These practices are frequently linked with soil organic carbon dynamics, microbial activity, and nitrogen use efficiency, highlighting an emerging interest in connecting emission outcomes with soil health and system sustainability. The co-occurrence of terms such as “carbon sequestration”, “crop residues”, and “crop rotation” suggests that mitigation is increasingly being evaluated alongside longer-term soil fertility and resilience considerations. Methane appears as a secondary theme, typically associated with soil moisture conditions and more complex or diversified production systems, such as agroforestry. This reflects the predominantly aerobic conditions of upland maize systems in SSA, where CH₄ emissions are generally low but may become important under specific hydrological conditions (Musafiri et al., 2020b).

Across the reviewed studies, maize-based agroecosystems in SSA exhibited distinct GHG emission characteristics shaped by low external inputs, strong seasonal climatic variability, and site-specific soil properties. Although CO₂, N₂O, and CH₄ were all reported, their relative contributions to overall climate impact differed substantially, with N₂O consistently dominating the global warming potential of maize systems (Table 2). Carbon dioxide emissions represented the largest absolute fluxes among the measured gases and were closely associated with soil respiration processes (Macharia et al., 2020; Musafiri et al., 2020b; Nyambo et al., 2020). Across studies, CO₂ fluxes displayed pronounced temporal variability, with emission peaks commonly occurring following rainfall events that stimulated microbial activity and root respiration (Maccarthy et al., 2018). Tillage intensity and residue management further influenced CO₂ dynamics, with higher emissions generally reported under conventional tillage and residue incorporation compared with reduced or no-tillage systems (Kimaro et al., 2016a; Nyambo et al., 2020; Muzangwa et al., 2021). Importantly, several studies reported increases in soil organic carbon under reduced disturbance and residue retention, indicating that short-term CO₂ emission pulses do not necessarily imply long-term carbon losses. Nitrous oxide was the most frequently measured greenhouse gas and emerged as the primary contributor to climate forcing in maize systems across SSA. Emission dynamics were strongly regulated by nitrogen availability and soil moisture, with pronounced N₂O pulses following fertiliser application and during wetting events after dry periods (Hickman et al., 2014; Zheng et al., 2019; Musafiri et al., 2020a). Across studies, N₂O responses to increasing nitrogen inputs were often nonlinear, with disproportionately large emission increases observed beyond certain fertiliser thresholds (Hickman et al., 2017; Zheng et al., 2019). Despite this sensitivity, absolute N₂O emissions reported in SSA were generally lower than those observed in high-input maize systems elsewhere, reflecting lower fertiliser use and nutrient-limited conditions typical of smallholder production systems (Leitner et al., 2020). For instance, N₂O emissions from maize farming systems in Michigan, USA, were found to be ten times higher than those in SSA and in Peru, maize-legume rotations receiving 100 kg N/ha had an emission factor of 1.53%, while a maize chop-and-mulch system in Brazil reported an emission factor of 2.3% (Pauw et al., 2018). Methane fluxes on the other hand, were assessed less frequently, consistent with the predominantly aerobic conditions of upland maize systems. Where measured, soils commonly functioned as net CH₄ sinks, particularly during dry periods. Episodic CH₄ emissions were observed under prolonged soil saturation, when anaerobic microsites favoured methanogenic activity and reduced CH₄ oxidation (Kimaro et al., 2016b; Ortiz-Gonzalo et al., 2018; Macharia et al., 2020; Musafiri et al., 2020a; Kibet et al., 2022; Mosongo et al., 2022). These findings indicate that CH₄ dynamics in maize systems are highly sensitive to short-term hydrological conditions and may become increasingly variable under changing rainfall regimes.

Conservation agriculture, based on the principles of reduced soil disturbance, permanent soil cover, and crop diversification, is widely promoted in SSA as a climate-smart approach capable of enhancing productivity while reducing GHG emissions (Kimaro et al., 2016a). Evidence from the reviewed studies indicates that CA practices can influence emission dynamics through multiple interacting pathways, but their mitigation outcomes are strongly context dependent and often involve trade-offs among different gases and management objectives (Kimaro et al., 2016a; Nyambo et al., 2020; Muzangwa et al., 2021; Vilakazi et al., 2021). Across studies, reduced or no-tillage systems consistently improved soil organic carbon retention and reduced CO₂ emissions relative to conventional tillage, particularly over longer time frames. Field experiments in South Africa and Tanzania, for example, reported lower CO₂ fluxes and higher soil carbon stocks under reduced tillage, reflecting reduced soil disturbance and slower organic matter decomposition (Kimaro et al., 2016b; Nyambo et al., 2020; Muzangwa et al., 2021). However, these benefits were not uniform across all conditions. In wetter environments or during prolonged rainfall periods, reduced tillage occasionally increased N₂O emissions due to higher soil moisture and the formation of anaerobic microsites that favoured denitrification (Nyambo et al., 2020). These findings highlight a key trade-off in CA systems, where practices that promote carbon sequestration may simultaneously increase short-term N₂O losses under certain environmental conditions. Residue retention, another core CA principle, further illustrates these trade-offs. Retaining crop residues generally enhanced soil moisture conservation and contributed to long-term soil carbon accumulation, but several studies reported short-term increases in CO₂ and N₂O emissions following residue addition, particularly when residues were combined with nitrogen fertiliser. For example, maize stover retention in Tanzanian systems (Centre and Centre, 2008) increased N₂O emissions during wet periods, while low-quality residues with high carbon-to-nitrogen ratios in other SSA systems promoted nitrogen immobilisation and reduced N₂O losses. These contrasting outcomes highlight the importance of residue quality, soil texture, and fertiliser management in determining emission responses. Crop diversification through rotations or intercropping influenced CA mitigation outcomes by altering nitrogen inputs and residue dynamics. While legume intercropping increased N₂O emissions in some studies due to biological nitrogen fixation, rotational systems incorporating legumes generally improved yields and reduced overall emissions compared with continuous maize monocropping. Long-term rotation trials in southern Africa demonstrated lower global warming potential and improved yield-scaled emissions under diversified systems, suggesting that rotation-based diversification may offer more consistent mitigation benefits than intercropping alone. Importantly, when mitigation outcomes were evaluated using yield-scaled metrics rather than absolute emissions, CA systems frequently performed better than conventional systems. Studies in Tanzania and South Africa showed that although absolute N₂O emissions were sometimes higher under CA practices, improved yields and soil carbon sequestration resulted in lower emission intensity per unit of grain produced (Chikowo et al., 2007; Atakora et al., 2019; Zheng et al., 2019; Fatumah et al., 2021; Tegha et al., 2024; Alasinrin et al., 2025). This finding emphasises that assessments of CA mitigation potential should account for productivity outcomes alongside emission measurements. Overall, the reviewed evidence indicates that CA holds significant potential to contribute to GHG mitigation in maize-based systems of SSA, but its effectiveness depends on the specific combination of practices implemented and their alignment with local soil and climatic conditions. Rather than delivering uniform emission reductions, CA practices influence emissions through complex interactions that require site-specific management and long-term evaluation. These insights reinforce the need to assess CA systems holistically, integrating emissions, soil carbon dynamics, and productivity to identify context-appropriate climate-smart strategies.

Despite the growing body of research on GHG emissions from maize-based systems in SSA, important evidence gaps and synthesis challenges remain. These gaps do not reflect a lack of agronomic or environmental research in the region, but rather the fragmented nature of existing studies, which differ widely in spatial coverage, experimental duration, measurement scope, and methodological approaches. A key limitation of the current evidence base is the strong geographic concentration of field-based GHG studies in a small number of countries, particularly in East and Southern Africa. Large maize-producing regions in West and Central Africa remain underrepresented in experimental emission measurements, limiting the ability to generalise findings across the region’s diverse agroecological zones. Expanding field-based measurements to these underrepresented regions would substantially improve regional emission estimates and mitigation assessments. Temporal limitations also constrain synthesis. Most studies are short-term, often spanning one to three growing seasons, and therefore capture only a limited range of climatic variability. Given the strong influence of rainfall, soil moisture, and temperature on emission dynamics, longer-term experiments are required to assess interannual variability and the persistence of mitigation outcomes under changing climate conditions. While individual components of conservation agriculture have demonstrated mitigation potential, their combined and long-term effects remain insufficiently quantified across diverse SSA environments. Addressing these gaps will require coordinated, multi-site, long-term experiments that integrate biophysical measurements with productivity and soil health indicators. Such efforts are essential for developing context-specific, climate-smart maize production strategies that balance food security and greenhouse gas mitigation objectives in Sub-Saharan Africa.