Conservation agriculture (CA) is widely promoted for its potential to support agroecological intensification in sub-Saharan Africa, yet adoption among smallholder farmers remains limited, partly due to the high labour demands of manual and conventional animal draught (CAD) practices. Mechanised conservation agriculture (MCA), enabled by low-cost two-wheel tractor technologies, offers a promising solution, though comparative field evidence under smallholder conditions remains scarce. This study investigated the agronomic and labour performance of MCA systems relative to CAD-based practices, focusing on maize and common bean production in Zimbabwe. On-farm trials were conducted across 94 farms over two consecutive seasons (2018-2019 and 2019-2020) under typical smallholder management conditions. MCA significantly outperformed CAD practices across both crops. Grain yields increased by 1.26 t/ha (66%) for maize and 0.9 t/ha (82%) for common bean (P < 0.001). Labour requirements were reduced by 124.6 person-hours/ha (a 5.55-fold reduction) for maize and 99.7 person-hours/ha (a 3.34-fold reduction) for common bean under MCA (P < 0.001). Agronomic indicators (including crop emergence, grain yield, and labour inputs) were assessed using ANOVA and generalized linear mixed models (GLMMs) to account for variability across farms and treatments. Fuel consumption under MCA averaged 5.2 L/ha for maize and 7.1 L/ha for common bean, representing a recurrent cost. However, the value of labour saved relative to fuel expenditure favoured MCA by approximately 8:1 for maize and 5:1 for common bean, indicating strong economic viability. Capital costs of equipment nonetheless remain a barrier to individual ownership. Study limitations include its short duration and context-specific conditions, which may constrain generalisability. Long-term, multi-location research across diverse agroecological zones is needed to further validate these findings. This study provides robust field-based evidence that MCA can substantially reduce drudgery, improve crop productivity, and enhance livelihoods in African smallholder farming systems.

Balmford and colleagues recently argued that yield maximization and land sparing are key to reducing biodiversity loss. Complementing the response from Beillouin and colleagues, this paper critiques such yield-centred approaches from an agricultural and livelihood perspective, focusing on smallholder-dominated systems in sub-Saharan Africa. In such contexts, increasing yields is often unprofitable, not the primary objective of smallholders, erodes ecosystem services and generates environmental spill-overs if implemented in isolation. Efforts targeting high-yielding regions are also likely to fail to address food insecurity, and to exacerbate inequities instead. Based on these insights, principles for biodiversity-friendly farming that account for smallholder realities and mediate trade-offs between production/livelihoods and conservation are outlined.

CONTEXT: Improved farming practices could contribute to solving food insecurity and poverty among smallholder farmers in Zimbabwe, who are dominantly mixed crop-livestock farmers. Agroecological practices, which rely on the use and recycling of local resources to reduce dependency on costly external inputs, are increasingly being claimed as suitable options for resource-constrained farmers. OBJECTIVE: The objective of this study was to assess how the use of improved crop and livestock practices, including agroecological ones, varied among different farm types. We hypothesised that agroecological practices would be preferentially used by farms with lower resource-endowment. METHODS: Data from a household survey of 605 farms across a representative sub-humid district and a representative semi-arid district in Zimbabwe, documenting the use of 32 common improved crop and livestock practices, 14 of which were classified as agroecological, were used for this study. A farm typology based on practice use was delineated using Multiple Correspondence Analysis, after which the socio-economic characteristics of the farm types and their associations with farm performance were assessed. RESULTS AND CONCLUSIONS: Our analysis revealed that farmers using a broader range of improved practices-including agroecological ones-were generally the wealthier and more resource-endowed households. Importantly, we found that these same farms were also more diverse in crop and livestock species, indicating that farm diversity was positively associated with productive capacity and resource endowment, rather than being a compensatory strategy of poorer households. Furthermore, we observed limited and inconsistent associations between the use of improved practices and production or diet outcomes. SIGNIFICANCE: The study points to the need for targeted support to less-endowed households, addressing not only technical but also socio-economic and institutional constraints to diversification and the use of improved practices.

Summary Intercropping cereals and legumes may enhance system productivity by simultaneously producing grain for human consumption and crop residues for livestock feed. This study evaluated the effects of maize ( Zea mays ) intercropped with cowpea ( Vigna unguiculata ) or mucuna ( Mucuna pruriens ) on soil moisture dynamics, grain and forage yields, and household food and feed self-sufficiency under varying rainfall conditions. The experiment was conducted on five farms in Mutoko District, Zimbabwe, during the 2021/22 and 2022/23 growing seasons using a Randomized Complete Block Design with five treatments: sole maize, sole cowpea, sole mucuna, maize–cowpea intercropping, and maize–mucuna intercropping. Soil moisture was measured biweekly to 1.8 m depth. Each season, experimental sites were classified according to seasonal rainfall as dry (<450 mm), average (450–600 mm), or wet seasons (>600 mm).Intercropping performance was influenced by rainfall. Sole mucuna utilized more soil moisture than other cropping systems. In dry and average seasons, sole maize yielded 1,812 kg ha −1 of grain, 103% more than maize intercrops, and met annual household maize requirements. In wet seasons, maize–cowpea intercropping produced higher maize grain yields (3,273 kg ha −1 ) than sole maize (2,372 kg ha −1 ), with a land equivalent ratio of 2.18. Mid-season dry spells reduced grain yields across treatments but had limited effects on stover production. During wet seasons, mucuna-based systems generated the highest forage and crude protein yields, sufficient to feed five cattle during the dry season. These findings highlight that maize–legume intercropping can enhance food and feed production during relatively wet seasons

Soil acidity is considered one of the most pressing problems for crop production in Ethiopia. Yet, estimates of the total area of acidic soils, defined here as soils with a pH value below 5.5, and of the proportion of acidic soils affected by aluminium toxicity are inconsistent. The aims of this review are: (1) to highlight the main challenges of soil acidity for crop production, (2) to review the evidence of crop responses to lime and phosphorus (P) on acidic soils from field and pot trials from Ethiopia, (3) to review policy documents on acidic soils and extent of soil acidity in Ethiopia and (4) to propose an agenda for future research. Soils can be acidified either by natural weathering processes or suboptimal management during crop production. Soil acidity affects crops through its effects on nutrient availability (mainly P availability and basic cations) and toxicities of aluminium and manganese. A review of studies of crop response to lime and P on acidic soils showed that in most studies, lime alone did not improve crop productivity. In contrast, application of P or combinations of lime and P together improved crop productivity. Re-interpretation of old documents, soil maps and recent digital soil maps revealed that about 4 % of the total Ethiopian land mass suffers from aluminium toxicity. Ethiopia's policy on acidic soil aims to improve the agricultural lime supply and increase the use of lime on acidic soils for increased productivity and fertilizer use efficiency. However, after reviewing studies on acidic soil management that reported crop response and the limited coverage of acidic soils with aluminium toxicity, the available evidence suggests that most acidic soils in Ethiopia can be effectively managed by supplying additional nutrients. This suggests that the focus of research for development should be broadened to integrated soil fertility management, tailored to local conditions, where lime use is targeted to aluminium toxic, acidic soils.

IntroductionForests support crucial ecosystem services globally, and contribute to the livelihoods of millions of people, but are threatened by agricultural expansion and intensification. Yet the relationship between forest cover and agricultural intensification, and how it mediates people’s access to and use of resources has been seldom studied globally.MethodsUsing remote sensing imagery and household surveys, we analyzed the effects of forest cover change on forest access (visitation) and household use of resources along a gradient from lower to higher land use intensification across seven tropical landscapes in low and middle income countries.ResultsWe find that increased agricultural intensification lowered forest visitation in the high intensification zone compared to the low to medium intensification zones. Forest types and changes, such as stable forests, forest gains and edge density, positively impacted household use of forest-sourced food and construction materials and visitation, highlighting the importance of sustainable forest management.DiscussionOur results complement the evidence of forest cover change by looking at agricultural intensification zones, which reveal the site-specific trends of forest types, ownership and household demand for ecosystem services. We recommend integrated landscape approaches with context and site-specific solutions to address the variability of forest types and multiple management goals that can be achieved with agricultural production and equitable access to provisioning goods and services in complex multi-functional landscapes.

CONTEXT: Farm mechanization has re-emerged as a development priority across sub-Saharan Africa, yet mechanization investments remain constrained by limited resources and uncertainty about where they are most likely to succeed. Existing approaches also rely on assumptions regarding the structural drivers of mechanization that are seldom evaluated empirically.OBJECTIVE: The objective of this study was to develop a spatial framework to guide mechanization investments by integrating an ex-ante, hypothesis-based approach with an ex-post, data-driven approach, and to evaluate the framework in Zimbabwe.METHODS: Principal component analysis and hierarchical clustering were used to identify mechanization recommendation domains, from five structural drivers commonly identified in mechanization theory: farming system intensity, land:labour ratio, early growing season rainfall variability, value of crop production, and market access. Using data from a national survey of 6,936 farming households, random forest models predicted the occurrence of nine mechanization outcomes from the same drivers as predictors, and quantified their relative importance.RESULTS AND CONCLUSIONS: Three mechanization recommendation domains were identified from the five hypothesized structural drivers and were strongly supported by observed mechanization ownership. Across the nine mechanization outcomes, value of crop production and early growing season rainfall variability consistently emerged as the dominant drivers of mechanization ownership at local level, whereas land:labour ratio contributed comparatively little. Contrary to prevailing mechanization theory, one recommendation domain exhibited consistently low mechanization ownership despite comparatively favourable farming system intensity, value of crop production, market access and land:labour ratio, indicating that current theory cannot fully explain mechanization patterns in Zimbabwe.SIGNIFICANCE: By integrating ex-ante and ex-post approaches, the proposed framework provides a practical basis for guiding mechanization investments and offers a means to empirically evaluate and refine mechanization theory across African farming systems.

While conservation agriculture (CA) has been widely evaluated for its biogeochemical effects (e.g soil organic carbon sequestration and greenhouse gas emissions) for climate mitigation, its biogeophysical impacts related to changes in surface albedo remain understudied. This study assessed the biogeophysical effects of CA cropping systems with maize (Zea mays L.) in Zimbabwe. Measurements were conducted continuously over two cropping years at two long-term experiments with contrasting soil characteristics, on an abruptic Lixisol and on a xanthic Ferralsol. The dynamics of surface albedo, longwave radiation, leaf area index, soil moisture and temperature were monitored under three different treatments: conventional tillage (CT, tilled to similar to 15 cm), no-tillage (NT) and no-tillage with mulch (NTM, 2.5 t DM ha(-)(1)). Our results revealed that, on the Ferralsol, NT and NTM significantly (p < 0.05) increased mean annual albedo (0.17) relative to CT (0.16), resulting in a negative instantaneous radiative forcing (iRF) and indicating a net cooling effect. iRF was stronger in 2021/22 (NT: -0.83 +/- 0.17 W m(-2); NTM: -1.43 +/- 0.7 W m(-2)) than in (2)022/23 (NT: -0.43 +/- 0.09 W m(-2); NTM: -1.03 +/- 0.21 W m(-2)). Conversely, on the Lixisol, while NT increased surface albedo (0.27 vs. CT: 0.24), NTM significantly reduced albedo (0.23), causing positive iRF (warming). iRF was -3.34 +/- 0.69 W m(-2) and -2.78 +/- 0.77 W m(-2) for NT in the first and second cropping year, respectively, and increased from 1.14 +/- 0.21 W (-2) (2021/22) to 2.77 +/- 0.41 W m(-2) (2022/23) under NTM. Overall, our results suggest that the soil background albedo is an important site characteristic that needs to be considered and demonstrates the importance of considering biogeophysical effects when promoting practices of CA for climate change mitigation.

Societal Impact Statement Gender responsive and more socially inclusive breeding strategies are needed to ensure new crop varieties, which offer greater yields in an increasingly variable climate, meet the needs of a diverse range of smallholder farmers. Participatory varietal selection actively involves farmers in testing and selecting new varieties at the end of the breeding process. We evaluated the inclusivity of a participatory maize breeding program in Zimbabwe. Our analysis found that setting targets for women farmer participation ensured adequate representation, but participatory research should move beyond simple targets and ensure the inclusion of different types of women and men farmers. Summary There is growing interest in participatory varietal selection and gender‐responsive breeding in research and development initiatives. On‐farm testing is increasingly used to ensure that new varieties perform within the target environments. However, there are few established approaches for selecting host women and men farmers who reflect the diversity of the overall target population of smallholder farmers. This study sought to evaluate ex‐post if recruited farmers within a participatory breeding network in Zimbabwe were representative of the surveyed population and pilot an approach to developing comprehensive farm typologies to ensure more gender‐responsive and socially inclusive breeding. A sample of over 2,000 randomly selected women and men farmers, including those hosting breeding trials, were surveyed. A typology was constructed to group farms with similar characteristics associated with household demographics, maize production and resource endowments. This facilitated the subsample of trial‐hosting farmers characteristics to be compared with the broader typology. The distribution of farm types selected by extension agents to host trials closely reflected the distribution of farm types within the surveyed population. Two farm types associated with women household heads and three types associated with men‐headed households were identified, highlighting the heterogeneity within these groups. Other important factors of differentiation included farm assets, livestock ownership and maize production features. Sampling strategies that explicitly incorporate agronomic and socio‐economic diversity within the target population should be used in the selection of host women and men farmers for participatory research to ensure appropriate gender and social inclusion.

The drive to mechanize and modernize African agriculture is in high gear, making the need for empirical evidence to guide mechanization investments critical. This paper assesses the heterogenous and distributional correlates of using mechanization and rural livelihoods in Chegutu and Zvimba districts of Zimbabwe, where a private sector company had the largest sales of different machinery across the country between 2019 and 2021. We used a quantile regression estimator and measured livelihoods using farm and household revenues. Based on survey data from 988 randomly selected households, we found that adoption was associated with rising land/labor ratio, market access and wealth. The use of mechanization was associated with a median annual increase of USD 262 in revenue with a wide range from USD 103 at the 25th percentile to USD 2,900 at the 95th percentile per year. The largest revenue gains were associated with post-harvest and irrigation equipment use, and in the upper percentiles of the revenue distribution. These findings call for (i) wealth agnostic promotional efforts to ensure equitable mechanization benefits, (ii) better targeting of mechanization types to farmer needs, and (iii) concerted efforts to strengthen mechanization service provision models.

CONTEXT: In Zimbabwe, farm productivity of smallholder farmers practising mixed crop-livestock farming is hindered by climate variability, inadequate nutritious feeds, poor soil fertility, and resource trade-offs. Despite these challenges, positive deviants (PDs) within these communities achieve better outcomes using resources similar to those of other farmers. OBJECTIVE: This study sought to identify crop-livestock practices that enable PDs to outperform low-efficiency farms (LEFs) and to compare their farm productivity (energy output), nutrient quantities added to croplands, gross margins and return on investment (ROI) from crop production. METHODS: Data from a survey conducted in Mutoko and Buhera districts of Zimbabwe in 2021 were used to derive a farm typology per district and identify PDs and LEFs within farm types. Selected farms were subjected to detailed surveys to identify their specific practices. RESULTS AND CONCLUSIONS: Compared to LEFs, PD farmers achieved significantly greater crop productivity - by 86 %, 89 % and 28 % - and livestock productivity - by 156 %, 101 % and 136 % on better-off, average and poorly-resourced farms, respectively. PDs had larger cropping areas (on average 42 % more) and owned more livestock (39 % more TLUs) than LEFs, but this does not fully explain differences in productivity. PDs used more inputs (fertilizer, labour and others) for crop production than LEFs and added more carbon and nitrogen to their soils. In both districts, PDs consistently outperformed LEFs in gross margins and ROI. The differences in economic performance between PDs and LEFs were more pronounced among the better-off farmers. Key practices contributing to PDs' success included recommended fertilizer use, timely operations, livestock supplementary feeding, fodder production, and adherence to extension advice. Financial shortages for the purchase of seeds, fertilizers, and veterinary drugs and poor access to information are potential hindrances to the adoption of PD practices by LEF farmers. SIGNIFICANCE: The combination of a farm typology and the PD approach helped to tailor recommendations to farms differing in resource-endowment, based on successful practices implemented in the region.

Problem: Semi-arid regions of sub-Saharan Africa are characterized by highly variable rainfall and low inherent soil fertility. Maize-cowpea intercropping may offer the prospect of increasing and stabilizing crop productivity in these regions. However, the performance of such cropping systems often varies considerably in space and time. Objective: The main objective of the study was to understand how farmer context and rainfall variability influence the performance of maize-cowpea intercropping, using on-farm field experiments together with soil-crop model simulations to compute water and nitrogen stress. Methods: The data used in this study was generated from twelve on-farm trials during two cropping seasons (2021/22 and 2022/23) in semi-arid Zimbabwe. Three maize (Zea mays L.) varieties, one cowpea (Vigna unguiculata (L.) Walp.) variety and two cropping systems-either sole or intercropped-were tested. The STICS soil-crop model was parameterized to reproduce crop growth in the on-farm trials and compute water and nitrogen (N) stresses. Linear mixed-effects models were used to assess the impact of experimental treatments and simulated water and N stresses on intercropping performance. Results: The Partial Land Equivalent Ratio (pLER-the ratio of intercropped productivity over sole crop productivity) for maize and cowpea greatly varied across farms and crop types. Maize variety did not significantly impact the pLER of maize and cowpea. Water stress and nitrogen (N) stress simulated by the model were significant predictors of variations in pLER: maize pLER for aboveground biomass significantly decreased with increasing simulated water stress, and maize pLER for grain yield significantly decreased with increased simulated N stress. Yet, average LER remained above one, regardless of the water or N stress on maize, because of a greater contribution of cowpea to LER when water and N stress on maize was high. Late planting was found to exacerbate maize water stress, while low total nitrogen in the top soil was significantly correlated with maize nitrogen stress. Conclusion: Our study reveals that the production benefits of maize-cowpea intercropping can be maintained, in conditions of high water and nitrogen stress in multi-year and multi-location on-farm experiments. Implications: Our findings confirm the assumption that intercropping is a useful approach to intensify and stabilize grain and fodder production in smallholder mixed crop-livestock farming systems in semi-arid environments.

Smallholder maize production in sub-Saharan Africa is constrained by soil fertility decline and erratic rainfall, leading to low dietary diversity. A four-season on-farm study (2018/2019-2021/2022) was conducted with six farmers in Murehwa District, Zimbabwe, to evaluate the effects of maize-legume strip cropping on soil nutrient dynamics, crop yield, and total system nutritional yield. The experiment was a 2 x 6 factorial arranged in a split-plot randomized complete block design, comparing sole maize with maize strip-cropped with pigeonpea (Cajanus cajan (L.) Mills), cowpea (Vigna unguiculata L. Walp.), lablab (Lablab purpureus (L.) Sweet), velvet bean (Mucuna pruriens (L.) DC), and groundnut (Arachis hypogaea L.), with and without Brachiaria ssp. (cv. Mulato II (CIAT36087) grass border. Rainfall ranged from 495 to 1053 mm, influencing crop and soil responses. Strip cropping affected soil properties between the 2020/21 and 2021/22 seasons, notably pH, exchangeable potassium (K), and soil organic carbon (SOC). Maize + cowpea improved K (+65 %) and pH (+30 %), while maize + groundnut improved SOC (+5 %). Maize grain yield was strongly influenced by season and its interaction with the strip cropping system. Maize + velvet bean yielded 2.4 t ha(-1) in 2018/19 and 2020/21, although sole maize produced the highest overall yield (3.2 t ha(-1)). Maize + pigeonpea consistently produced the highest grain yield among legumes (approximate to 1.0 t ha(-1)) and total system protein yield (>0.3 t ha(-1)), while starch yield was highest under sole maize in 2020/21. Overall, maize-legume strip cropping, particularly pigeonpea, cowpea, and groundnut, improved system-level nutrient yield and selected soil properties, demonstrating its potential for climate-resilient smallholder farming.

Acid soils are widespread across sub-Saharan Africa. Agricultural lime can be used to alleviate production constraints associated with soil acidity, but lime is not widely available in the region, and it is unclear if applying it would be profitable. Using lime requirement models and crop yield responses to soil acidity modelled as plateau-linear decay functions, we estimated the profitability of acid soil remediation through liming. Crop yield loss to soil acidity occurs on 32.7 Mha, or 23% of sub-Saharan Africa's cropland. The burden of acid soils is US$6.0 billion (6% of the current production value), and 75% of that could be profitably alleviated. Under prevailing conditions, liming would be profitable in the year of application on 6.2 Mha (with an average profitability of US$278 ha-1) and on 8.8 Mha when lime's long-term effect is considered. Intensification of crop production and lower relative lime/output prices could make liming profitable on more cropland.

Societal Impact StatementA hybrid maize seed production technology has the potential to reduce the complexity of hybrid seed production and increase seed quality. Here, we investigate the potential impact of this technology on yields when hybrid maize is recycled. Hybrid maize recycling is a practice used by resource‐poor farmers as a coping mechanism during drought years. Recycling hybrid maize produced using this technology could provide a small yet significant yield benefit to resource‐poor farmers when they chose to recycle. This study provides an example of how social considerations can be incorporated into testing strategies of new technologies to ensure equitable benefits.Summary Understanding the performance of new genetic technologies in farmers' real‐world realities, especially those relevant to resource‐poor farmers, is often overlooked but is essential to ensure equitable benefits. A new genetic technology was developed to simplify hybrid maize seed production in sub‐Saharan Africa, thereby improving farmers' access to high‐quality hybrid seed. Hybrids produced with this technology segregate 50:50 for pollen‐producing and non‐pollen producing and are designated 50% non‐pollen producing (FNP). FNP maize has higher yields in low‐input environments. As recycling hybrid maize seed remains a common practice in Zimbabwe, including among resource‐poor households, it is important to understand the impact of recycling FNP seed on the yield gains from the FNP technology. The potential impact of recycling FNP hybrid seed was assessed by testing three seed recycling scenarios on‐station and on‐farm. The extent of hybrid seed recycling and the types of households recycling hybrid maize seed over a 3‐year period were also investigated. Hybrid maize seed recycling was associated with resource‐poor farmers, although it was not continually practiced across years. Yield gains associated with FNP were retained under recycling practices, albeit reduced. The greatest yield benefit was when seed from only non‐pollen‐producing plants was used. Yield gains were associated with longer ears and more kernels per ear. While recycling hybrid maize seed reduces potential yields due to inbreeding depression, in the years when farmers cannot afford to plant hybrid maize only, recycling non‐pollen‐producing hybrid maize seed conferred a yield benefit of 116 kg ha−1.

Soil acidity is a major constraint to crop production in tropical regions. Although agricultural lime is one option to remediate acid soils, there is limited information on the potential returns on investments to liming by smallholders. Using survey data collected from 261 households in Rwanda, we estimated the crop -specific yield response to lime application and associated financial benefits. The estimated average yield gain from lime ranged from 941 kg/ha to 1 579 kg/ha for Irish potato, 562 kg/ha to 709 kg/ha for maize, and 453 kg/ha to 520 kg/ha for beans. With the existing lime and farmgate crop prices, reliable returns on investment from lime were observed for Irish potato, while applying lime to maize and bean was only profitable at a 50% lime price subsidy. As maize and beans are the major staple crops in Rwanda, the subsidy for ag-lime use in improving crop productivity is highly justifiable. The results inform policy decisions in considering market -oriented crops and subsidies when promoting agricultural lime in acid soils under smallholder conditions.

Sustainable agricultural practices such as conservation agriculture have been promoted in southern Africa for nearly three decades, but their adoption remains low. It is of policy interest to unpack behavioural drivers of adoption to understand why adoption remains lower than anticipated. This paper assesses the effects of risk aversion and impatience on the extent and intensity of the adoption of conservation agriculture using panel data collected from 646 households in 2021 and 2022 in Zambia. We find that 12% and 18% of the smallholders were impatient and risk averse, respectively. There are two main empirical findings based on panel data Probit and Tobit models. First, on the extensive margin, being impatient is correlated with a decreased likelihood of adopting combined minimum-tillage (MT) and rotation by 2.9 percentage points and being risk averse is associated with a decreased propensity of adopting combined minimum tillage (MT) and mulching by 3.2 percentage points. Being risk averse is correlated with a decreased chance of adopting basins by 2.8 percentage points. Second, on the intensive margin, impatience and risk aversion are significantly correlated with reduced adoption intensity of basins, ripping, minimum tillage (MT), and combined MT and rotation by 0.02–0.22 ha. These findings imply a need to embed risk management (e.g., through crop yield insurance) in the scaling of sustainable agricultural practices to incentivise adoption. This can help to nudge initial adoption and to protect farmers from yield penalties that are common in experimentation stages.

An innovative methodological approach combining statistical typologies and stochastic frontier analysis was applied to data collected from 1840 mixed crop-livestock farms in six districts of Zimbabwe, representative of semi-arid areas of the country. The average annual cereal production was 362 kg farm-1, and the average annual livestock offtake was 0.64 +/- 1.32 Tropical Livestock Units (TLU) farm-1. Our results demonstrate there is scope to increase cereal and livestock production by 90.7% and 111.9% relative to current production levels, respectively, with more efficient use of existing resources and technologies. Rainfall was found to have a strong effect on cereal production, highlighting the need for climate-smart practices. Livestock mortality (0.59 +/- 1.62 TLU farm-1) was found to be in the same order of magnitude as livestock offtake (0.64 +/- 1.32 TLU farm-1). Cereal production was supported by livestock, demonstrating the importance of crop-livestock interactions in these mixed farming systems. Three farm types were identified in our analysis. Crop-oriented mixed farms (31%) are likely to be the ones most responsive to crop-specific interventions e.g., crop rotation and integrated pest management. Livestock-oriented mixed farms (34%) are likely to benefit the most from livestock-specific interventions, e.g., home feed. Mixed farms dependent on off-farm activities (36% of the sample) may require nutrition-sensitive and labour-saving sustainable intensification technologies to benefit from their limited resources. Reducing cattle mortality is a priority for all three farm types. The method proposed here could be adapted to other contexts characterized by heterogeneous farming populations to target interventions.

Background: Evidence of the effectiveness of biofortified fi ed maize with higher provitamin A (PVA) to address vitamin A deficiency fi ciency in rural Africa remains scant. Objectives: This study projects the impact of adopting PVA maize for a diversity of households in an area typical of rural Zimbabwe and models the cost and composition of diets adequate in vitamin A. Methods: Household-level weighed food records were generated from 30 rural households during a week in April and November 2021. Weekly household intakes were calculated, as well as indicative costs of diets using data from market surveys. The impact of PVA maize adoption was modeled assuming all maize products contained observed vitamin A concentrations. The composition and cost of the least expensive indicative diets adequate in vitamin A were calculated using linear programming. Results: Very few households would reach adequate intake of vitamin A with the consumption of PVA maize. However, from a current situation of 33%, 50%-70% - 70% of households were projected to reach >= 50% of their requirements (the target of PVA), even with the modest vitamin A concentrations achieved on-farm (mean of 28.3 mu g RAE per 100 g). This proportion would increase if higher concentrations recorded on-station were achieved. The estimated daily costs of current diets (mean f standard deviation) were USD 1.43 f 0.59 in the wet season and USD 0.96 f 0.40 in the dry season. By comparison, optimization models suggest that diets adequate in vitamin A could be achieved at daily costs of USD 0.97 and USD 0.79 in the wet and dry seasons, respectively. Conclusions: The adoption of PVA maize would bring a substantial improvement in vitamin A intake in rural Zimbabwe but should be combined with other interventions (e.g., diet diversification) fi cation) to fully address vitamin A deficiency. fi ciency.