Fertilizer and soil health for sustained crop production in sub-Saharan Africa: Theory and conceptual framework

The urgent need to increase agricultural production in sub-Saharan Africa (SSA) to ensure food and nutrition security for its growing population has been widely acknowledged^1,2. Pressure on land in SSA is aggravated by increasing global demand for agricultural products such as cocoa or oil palm³. If intensification of crop production is unsuccessful and massive cropland expansion is to be avoided, SSA will become much more reliant on cereal imports than it is today⁴. Despite the overall disappointing agricultural production statistics across the continent, some African countries have moved toward a trajectory similar to Asia, achieving higher crop yields on the same land area through the increased use of mineral fertilizer, henceforth referred to as ‘fertilizer’, sometimes with rates reaching the target of 50 kg of fertilizer nutrients per hectare, proposed by the Abuja 2006 Fertilizer Summit⁵ (Fig. 1).

Acknowledging that progress over the past two decades has been variable and that new, more comprehensive strategies are needed to improve fertilizer use and soil health in Africa, the African Fertilizer and Soil Health Summit was held in Nairobi in 2024. The summit culminated in the adoption of the Nairobi Declaration and a 10-year action plan⁶ to revitalize the concept of soil health and enhance agricultural productivity across Africa. The plan calls for more context-specific, integrated approaches to soil fertility management, embracing the principles of integrated soil fertility management (ISFM)⁷.

Soil degradation is a major factor underlying the low crop productivity and high prevalence of malnutrition in SSA⁸. Soil degradation in Africa is largely driven by suboptimal application and management of nutrients, resulting in a downward spiral of poor crop grain and biomass yields, resulting in less organic inputs and soil cover. Degradation triggers a decline in soil biological, chemical, and physical quality, and its capacity to support crop production and regulate environmental functions—together referred to as soil health⁹. Decades of soil nutrient mining—continuous cropping with few nutrient inputs has severely degraded the productive capacity of vast agricultural areas in SSA, making it the only global region with persistently negative nutrient balances that have worsened over time¹⁰ (Fig. 2). Increasing crop productivity to meet current and future food needs while preserving soil health will not be possible without increased and efficient use of all possible nutrients, with critical contributions from mineral fertilizers^4,11.

The main objectives of this paper are: (i) to illustrate the fundamental role of soil health in supporting sustainable agriculture; (ii) to review the evidence on how fertilizer use interacts with soil health; and (iii) to recommend how fertilizer can be used more efficiently to improve agricultural productivity and soil health. We focused on upland, rainfed annual cropping systems, since these are the most common farming systems managed by smallholder farmers in SSA (Table 1). While we recognize that the recommendations for these systems may not apply to irrigated lowland agriculture or perennial systems, most smallholder farmers in SSA and in other parts of the tropical regions, including those with part of their farm under perennials or lowland rice, rely heavily on these rainfed systems for their food requirements and livelihoods.

Description of functions and properties

Soil health is commonly defined as the continued capacity of soil to function as a vital living system, within ecosystem and land-use boundaries, to sustain biological productivity, promote the quality of air and water environments, and maintain plant, animal, and human health¹². In the context of agricultural soils, it refers to the capacity to support the production of food, feed, and fiber in amounts and quality sufficient to meet human and animal needs while delivering other ecosystem services that are essential for maintaining the quality of life for humans and the conservation of biodiversity¹³. To deliver those goods and services, soils perform a number of essential functions, including (i) regulation of water flows and storage, (ii) storage, regulation, and recycling of nutrients, (iii) provision of physical stability and support, (iv) provision of habitat and energy for soil organisms, (v) filtration and buffering of potential pollutants, and (vi) mitigation of climate change¹⁴. Several soil properties underpin these functions, classified as chemical, physical, and biological properties.

Soil chemical properties include total nutrient stocks, amount of available or exchangeable nutrients, acidity and buffering capacity, cation and anion capacity, soil organic matter content, base saturation, and salinity. Soil physical properties include color, texture, structure, porosity, bulk density, aggregate stability, and temperature. Soil biological properties include abundance, diversity, and activity of soil biota, including macro-, meso-, and microfauna, or proxies for the processes they perform in soil (e.g., carbon (C) or nitrogen (N) mineralization potential, and soil enzyme activities). Standard methods for assessing chemical and physical properties have been available for many decades. Further, the chemistry, physics, and biology of soils are interdependent—for example, macrofauna such as earthworms and termites are key to porosity. How these properties are affected by inherent soil conditions or management practices and their interpretation in relation to specific soil functions is relatively well understood^15,16. Despite the importance of biological properties for specific soil functions, biological soil health indicators have been underutilized, due to the complexity of the soil biotic communities and challenges with establishing causal relationships between biological properties and specific functions⁹ except for keystone species linked to specific soil processes or functions^16,17. Future research needs to elucidate the role of soil biota in soil health assessment, including the support of soil functions for production and ecosystem services.

Because of the diverse functions regulated by soil and the large number of soil properties that interact with those functions, a practical indicator framework for soil health needs to be constructed based on indicators that have causal relationships with key functions that can be measured over time and space¹⁸. For instance, while the critical soil functions in a natural ecosystem do not include sustaining crop production, this function is obviously of utmost importance for agroecosystems. Hence, several soil indicators and their critical values in natural ecosystems are, by definition, not valid targets for healthy soils under agriculture.

Indicators of soil health from an agronomic perspective

Here, we focus on soil health in support of crop production, which requires, besides light and carbon dioxide, sufficient and timely access to water, a sufficient and timely amount of nutrients with the right composition, and proper root development to recover nutrients and water. It also requires sufficiently low levels of toxic elements that may hamper plant growth, such as exchangeable aluminum (Al), or soil-based pests and diseases, such as root knot nematodes.

We suggest that the key indicators to sustain crop production are soil organic matter (SOM), available P and exchangeable K, soil acidity, expressed through its pH, and bulk density. Available N is highly dynamic in the soil, making it an unreliable indicator, but SOM provides a good proxy for total N. Common C:N:P ratios of agricultural soil are around 100:10:1¹⁹, and 1 ton of SOM contains about 600 kg of C, 60 kg of N, and 6 kg of P, all gradually released through decomposition. Soil organic matter is not only a source of nutrients but also a temporary store through its cation exchange properties²⁰ that also determine the proton buffering capacity of SOM. Additionally, SOM influences soil structure and thereby bulk density and soil porosity and thus water storage, movement, and root growth²¹. The amount of SOM and the potential of soil to accumulate SOM are strongly influenced by the amount and reactivity of the clay (and silt) particles in the soil²². As such, determination of critical thresholds for SOM content in relation to soil health must take into consideration specific soil texture and mineralogy classes.

Phosphorus (P) and potassium (K) are the most limiting plant nutrients besides N and are required by all crops in substantial quantities, with the root and tuber crops, as well as bananas, requiring particularly large amounts of K. Moreover, for legumes to effectively fix considerable amounts of N from the atmosphere, sufficient P is required²³. Soil acidity influences the availability of nutrients with direct impact on crop growth ²⁴ and regulates the amount of exchangeable Al or manganese (Mn), which are toxic and hinder crop growth²⁵.

Soil bulk density is a critical physical property influencing root growth, water infiltration and retention, and soil aeration. For example, high bulk density indicates compacted soil, which restricts root penetration and expansion. It can also lead to reduced water infiltration and increased surface runoff, which limits water availability to plants. At the same time, high bulk density will also limit nutrient availability by inducing anaerobic conditions, which will reduce nutrient mineralization and favor gaseous losses of N as N₂O and N₂²⁶.

Thresholds for soil properties can be broadly categorized into three types: ‘killer’ thresholds, where exceeding a critical limit causes immediate and severe loss of function; ‘soft’ thresholds, where functionality declines gradually once crossed; and ‘more is better’ thresholds, where increasing levels consistently improve the soil function without a clearly defined upper or lower limit. For example, on soils with substantial Al contents, soil acidity has a killer threshold at pH values of 5.5, below which Al saturation can exceed 15%, thus hampering crop growth²⁷. At the other extreme, soil pH values above 9 limit the availability of specific nutrients such as P, Mn, Zn, or B²⁸. Available P and exchangeable K have soft thresholds above which there are no limitations to crop growth, and below which crop growth is reduced but not ceased. As for SOM, there is no easily identifiable minimum threshold, but in general, more SOM is better for soil health and crop productivity.

In addition to the need for specific indicators to assess soil health, three other key factors must be considered in a soil health indicator framework: (i) spatial heterogeneity—some soil properties can vary significantly over short distances, particularly in densely populated areas with absence of fallow land²⁹; (ii) rate of change—some indicators change rapidly (e.g., changes in available nutrients), while others change slowly, especially when these are measured against a relatively large background (e.g., changes in SOM); and (iii) measurement costs— assessing indicators over large areas of land can be expensive although the availability of spectroscopic methods have succeeded in substantial reductions in cost for some soil health indicators. As an alternative to direct measurement, changes in crop and soil management practices can be evaluated as proxies for expected changes in key soil health indicators through empirical or mechanistic modeling approaches³⁰.

Finally, much of SSA is characterized by highly weathered soils, which often result in more than one nutrient deficiency, a contrast to most soils in the northern hemisphere. Varying proportions of the soils in SSA suffer from secondary and micronutrient limitations, including S or Zn³¹ and this can be exacerbated when focusing solely on NPK fertilizer. Obviously, addressing these location-specific deficiencies is crucial for resolving soil health and enhancing crop productivity.

The conceptual framework relating fertilizer use to soil health includes the main components of cropping systems and their interactions, along with management interventions and inherent soil properties (Fig. 3). Crop production levels are the result of yield-defining (radiation, temperature, and CO₂), yield-limiting (water, nutrients), and yield-reducing factors (biotic factors, pollutants)³². Soil health influences both yield-limiting factors (water (oxygen) and nutrients) and yield reducing factors (pH, pest/diseases and pollutants), while fertilizer and lime can (i) alleviate nutrient-limiting factors, (ii) increase crop yields and thus residues and SOM or soil health, and (iii) directly affect soil health (e.g., through increased nutrient availability and influences on soil pH).

Nutrients can be supplied either from fertilizers, manure, compost, crop residues, biological N fixation or derived from soil nutrient stocks in the form of SOM or weatherable minerals. Water availability, either through rainfall or irrigation, and soil aeration status mirror each other, and the correct balance between the two is affected by soil structure, which is influenced by crop residue or manure application and land preparation^33,34. Soil organic matter is increased with the application of crop residues, either directly or after passage through livestock as manure³⁵.

The above processes are affected by specific inherent soil properties, affected by their position in a landscape. Soil depth, limited by physical barriers such as hardpans or a high water table, affects the soil volume that plant roots can explore for nutrients and water. Soil texture affects the potential of soils to form soil structure, with sandy soils having limited scope for the formation of a stable structure, while clay type affects the potential to bind organic matter, with 1:1 clays having less potential than 2:1 clays^22,36. In general, soils in SSA are mostly highly weathered and dominated by 1:1 clays with limited capacity to store both organic matter and nutrients in an exchangeable form³⁷. Additionally, the reserve of weatherable minerals in many soils of SSA is small³⁷ constituting a major difference with the much younger soils in the northern hemisphere.

Fertilizer use and crop yields

National-level data and other evidence, such as soil nutrient balances, show that fertilizer use is necessary to increase and sustain crop yields in SSA^11,38,39. There are many ways to assess fertilizer use efficiency (FUE)⁴⁰ each with its own requirements and limitations⁴¹. Among these, agronomic use efficiency (AE) is a commonly used indicator due to its ease of assessment. Agronomic use efficiency of N (AE-N) is calculated as (Eq. 1):

However, the assessment of AE requires data on crop performance in the absence of fertilizer (control treatment), which is often missing in surveys, national statistical databases, and farmer trials. In such cases, partial factor productivity (PFP) of fertilizer (Eq. 2) is used as an alternative, estimated as the ratio of yield to fertilizer nutrients applied, for instance, for N fertilizer:

Other approaches for estimating FUE exist, but they typically require direct measurements of nutrient uptake, limiting their use to experimental field studies.

African farmers generally achieve highly variable and relatively low FUE values, resulting in marginally profitable or unprofitable fertilizer use^42,43. In contrast, researchers’ experiments consistently report substantially higher FUE values⁴⁴. To maximize FUE, fertilizer application must follow the ‘4 R nutrient stewardship’ principles—applying the right source, at the right rate, at the right time, and in the right place⁴⁵. Moreover, other management practices and decisions are often equally important to maximize FUE, including the choice of germplasm and planting dates, the combined application with organic inputs, the deployment of other amendments (e.g., lime to address soil acidity-related constraints), or other practices (e.g., weed, pest, and disease management or sub-surface plowing in case of hardpans), as conceptualized through the ISFM paradigm^7,8.

There is ample evidence that maintaining and improving SOC leads to higher FUE. A clear example comes from a 3-year study on millet^35,46, which found significantly higher FUE in the more fertile fields near the farm homestead compared to the less fertile fields further away. Additionally, integrating mineral fertilizers with organic inputs, as part of ISFM, has been shown to improve FUE by alleviating key nutrient uptake constraints³⁵. In general, the effects of mineral fertilizer and organic inputs are additive with moderate application rates, while negative interactions between both inputs are exceptional⁴⁷.

Fertilizer use and soil health

Fertilizer application can affect soil health in both positive and negative ways. For example: (i) application of P and K-based fertilizers can increase stocks of available P and K beyond the season of application, or (ii) fertilizers can increase crop yields and thus allow return of larger amounts of crop residues to the soil, thereby potentially enhancing SOM content and quality. Furthermore, ammonium-based fertilizer (e.g., urea, ammonium sulfate) can gradually increase soil acidity through the release of protons during the nitrification process.

When P fertilizers are applied to soil, a portion of the P is adsorbed onto soil minerals or precipitated into less soluble forms. Over time, some of this sorbed P can desorb back into the soil solution, becoming available for plant uptake in subsequent seasons. The desorption rate depends on soil properties such as pH, clay content, and organic matter⁴⁸. Several studies in SSA have shown that both natural rock phosphate and single superphosphate fertilizers have significant residual or ‘legacy’ effects, improving soil fertility and increasing cereal crop yields⁴⁹. Potassium applied as fertilizer is more mobile in the soil than P but leaches less than nitrate, except in sandy or highly weathered soils⁵⁰. In medium- and heavy-textured soils, a significant portion of applied K remains in the root zone and is available for subsequent crops. In soils with illite, vermiculite, or smectite clays, K+ ions can be trapped between clay layers. Over time, as soil moisture and temperature fluctuate, some of the trapped K⁺ slowly re-enters the soil solution, making it available for plant uptake in subsequent seasons⁵¹.

Theoretically, fertilizer use can enhance SOC by boosting crop plant productivity and thereby increasing C inputs to the soil through greater biomass returns⁵². However, a review of published meta-analyses on the effect of mineral fertilizer on SOM⁵³ found only one meta-analysis covering experiments in SSA⁵⁴, which showed no significant effects of fertilizer application on SOM (Fig. 4). This could be the result of low fertilizer application rates and/or sub-optimal associated agronomic factors, further aggravated by the removal of crop residues in many African farming systems for alternative uses. Similarly, results from four long-term experiments in Kenya^35,55 and one in the Ivory Coast⁵⁶ indicate that while mineral N fertilizer is crucial for short-term maize yield improvements, its long-term use without organic inputs does not increase SOC. On the other hand, Zingore et al.⁵⁷ found that commercial farmers in Zimbabwe, through the use of mineral fertilizers and hybrid maize varieties, and recycling of crop residue, maintained SOC contents almost double those in communal farming systems on similar soils without external inputs. Ultimately, it is likely that to increase SOC, both internally generated (e.g., crop residues) and external organic inputs (e.g., manure, compost) may be required. Options also exist to ‘internalize’ the production of organic matter, e.g., through agroforestry options or herbaceous legume intercrops or rotations, but the uptake and scalability of such interventions face other challenges beyond the scope of this paper⁵⁸.

Under unbalanced fertilization, some studies have shown that mineral fertilizers can reduce SOC through priming effects in the rhizosphere—a process whereby plants and microbes decompose native SOM to access limiting nutrients⁵⁹. This phenomenon is often driven by stoichiometric imbalances, especially in micronutrients, prompting microbial mining of SOM to satisfy nutritional demands.

A specific concern of soil health for SSA is soil acidity⁶⁰. The dominance of weathered soils in SSA implies a small reserve of weatherable minerals and a limited buffering capacity against pH changes⁶¹. Given the current limited use of lime, care should be taken when selecting the type of fertilizer, as the application of ammonium-containing fertilizers may aggravate the situation²⁹. Ammonium-based fertilizers (e.g., urea, ammonium sulfate) increase soil acidity over time through the process of nitrification, which releases protons (H⁺) in the soil solution. Additionally, H⁺ displaces cations (Ca²⁺, Mg²⁺, ānd K⁺), which are subsequently leached, further acidifying the soil. Over time, increased acidity can lead to aluminum (Al³⁺) and manganese (Mn²⁺) toxicity, harming plant roots⁶². Application of lime can mitigate this acidifying effect, but its economic and agronomic effects need a long-term perspective⁶³.

Crop productivity and soil health

Increased crop productivity generates larger amounts of crop residues, including roots, which can enhance soil health. On average, a 1.3% annual increase in crop grain yields is associated with a 0.4% annual increase in SOC, based on studies from Africa, Asia, and Latin America (Fig. 5). The large scatter in the data results indicates that several other factors besides increased crop productivity affect changes in SOC, suggesting the importance of additional management practices to ensure that increased crop yields and biomass productivity have a positive effect on soil health. Several studies⁵⁵ have shown that farmyard manure is more effective in stabilizing SOC than fresh crop residues. Farmyard manure contains partially decomposed organic matter, including humified compounds that are more resistant to microbial breakdown, also forming stable soil aggregates that protect SOC from decomposition²².

Moinet et al.⁶⁴ reviewed 21 meta-analyses and found that the effects of increasing SOC on crop yields are highly variable, ranging from negative to neutral to positive. SOC sequestration does not universally improve yields, and any positive effects are highly context dependent. Factors like soil texture, climate, and nutrient availability significantly influence yield responses, and establishing causality between SOC stocks and crop yields is therefore challenging. Agricultural practices that increase SOC may also improve yields, meaning their correlation over time does not necessarily imply a direct causal relationship. Even if a causal link exists, it is difficult to determine whether SOC influences yield or vice versa. Calibrated crop models tailored to specific conditions could help address this uncertainty⁵².

Crop residues help mitigate this by returning essential cations like calcium (Ca²⁺), magnesium (Mg²⁺), and potassium (K⁺) to the soil as they decompose. This also helps neutralize soil acidity by replacing hydrogen (H⁺) and aluminum (Al³⁺) ions on soil exchange sites. However, when crops remove more cations (Ca²⁺, Mg²⁺, K⁺) than those replenished through residue decomposition, soil acidification may occur over time⁶⁵.

Integrated soil fertility management

The co-application of mineral fertilizers with organic inputs is a foundational principle of ISFM⁷ because neither input alone can sustainably support crop productivity and soil health in smallholder systems of SSA. Organic inputs such as crop residues, manure, and compost increase SOC, contribute to improved soil structure, water retention, cation exchange capacity, and biological activity, and help buffer soils against acidity and nutrient losses. However, when applied at quantities typically available on smallholder farms, they rarely release sufficient nutrients to meet crop demand¹¹. Mineral fertilizers, in contrast, provide concentrated and immediately available nutrients that enable rapid yield increases but are most effective when soil physical and biological constraints are alleviated through organic matter management.

Evidence from ISFM research shows that positive interactions between mineral and organic nutrient sources frequently enhance agronomic use efficiency by improving nutrient capture and conversion efficiency, particularly on degraded or poorly responsive soils⁵⁴. Fertilizer application can also stimulate biomass production, thereby increasing the availability of crop residues that could have a positive impact on SOC and protect soils against erosion, if produced and retained in sufficient quantity. On the other hand, sustained application of organic inputs can rehabilitate degraded soils and restore their responsiveness to mineral fertilizers. These complementary effects demonstrate that integrated management of mineral fertilizers and organic resources is essential not only for maximizing short-term yield responses but also for rebuilding soil fertility and ensuring long-term sustainability of cropping systems in SSA.

Although fertilizers are a key component of sustainable food systems and essential for maintaining soil health, their widespread use in Africa heavily relies on policy support⁶⁶. Adoption and application rates among smallholder farmers in SSA remain far below optimal, despite many initiatives to promote fertilizer use⁶⁷. The prime cause for low adoption of fertilizer is its high cost relative to prices that farmers get for their produce. Crop response to sole application of fertilizers is poor without complementary good agronomic practices, leading to low, highly variable, and even negative economic viability⁴³. In addition, a range of factors influence fertilizer use, including household and farm characteristics, social and human capital, type of crops produced, credit access, off-farm income, regular labor availability, and farmers’ perception of the effects of fertilizers on soil fertility⁶⁸. Moreover, biophysical conditions, such as rainfall patterns and soil type, are important; in particular, the high risk of crop failure resulting from low or erratic rainfall is a strong disincentive to investing in fertilizers for subsistence crops⁶⁹.

Crop yields in SSA remain well below what is attainable given the environmental conditions and available technologies, referred to as the ‘agronomic’ yield gap. Even though it may be technically feasible to reduce yield gaps in many regions of SSA, it is not always economically viable for farmers to pursue higher crop yields. For example, Bonilla-Cedrez et al.⁷⁰ showed for maize that (i) the average economic yield gap (the difference between current yield and those that would maximize profit) was about 25% of the agronomic yield gap; (ii) although maize yields could, in theory, be profitably doubled, actual economic incentives are weak due to unfavorable and highly variable input-to-output price ratios; and (iii) risks associated with erratic seasonal rainfall further discourage investment in fertilizer.

Farmers who grow high-value crops or crops under irrigation are more likely to invest in fertilizers and other soil health-enhancing management practices as a way to increase profitability. While investments are generally lower for traditional staple food crops, rising food market prices, often driven by their use as raw materials in processing industries (e.g., cassava for flour), can encourage farmers to adopt fertilizers. Land is a key factor of production, and farmers with larger farm sizes are generally more inclined to invest in fertilizers and soil health-enhancing practices. In addition, secure land tenure (whether in terms of access, use, or ownership) also encourages investment in fertilizer and other technologies, in contrast to situations of insecure tenure, which tend to discourage long-term investment in cropland^71,72. Furthermore, land ownership often enables access to credit, an important factor for investments in fertilizers.

Farmers’ age, gender, and education level are important mediating factors that shape their knowledge, ability, decision-making capacity, preferences, and risk tolerance, all of which influence the adoption of agricultural technologies. Lower adoption of fertilizer use by women farmers may be attributed to differences in technology preferences, cultural norms, and the suitability of agricultural tasks typically performed by women⁷³. Furthermore, women’s adoption of technologies is frequently constrained by access to inputs, extension support, and access to credit and markets. They may also lack complementary resources, especially land and labor.

There is a broad consensus that agricultural productivity in SSA can and must increase to meet food needs, while expansion of arable land, deforestation, and loss of biodiversity must be avoided. Achieving this requires fertilizer and organic inputs, in combination with good agronomic practices, improved seeds, and other amendments, core components of the ISFM approach. Building soil health or regenerating degraded soils is a long-term process, while the loss of soil health is rapid when nutrients and SOC are not replenished. Building soil health will not be possible without the use of fertilizers, as available organic inputs provide only a part of the required nutrients, and their transfer from one location to another leads to soil depletion and degradation elsewhere.

A major barrier to the adoption of soil health-restoring practices by smallholder farmers is the time lag before visible benefits appear. In the absence of incentives and driven by short-term economic rationale, farmers tend to favor interventions that yield immediate returns. With the broadening of soil fertility management recommendations beyond mineral fertilizers, practical tools that translate complex information into actionable guidance are needed. Digitally enabled tools, together with scaling models and bundled services, that include good agronomic practices with socio-economic drivers like credit access, can help bridge this gap, facilitating farmers’ access to balanced sets of agro-inputs, advice, and technologies. Policy interventions are also essential, creating an enabling environment for the adoption of fertilizer and soil health management recommendations, including institutional incentives, smart subsidies, and knowledge transfer through an effective extension system. In this context, it is crucial to redesign fertilizer subsidy programs beyond reducing cost toward actively promoting soil health outcomes.

Instead of a blueprint approach, soil health recommendations should always be tailored to local conditions. Several approaches can be used to develop optimal fertilizer recommendations that account for crop nutrient requirements and field-specific soil characteristics. This involves assessing the soil fertility status of individual fields. New data and analytical tools now allow for the development of site-specific recommendations at scale. In particular, recent tools based on artificial intelligence can improve the development and delivery of hyper-local advisory services tailored to a single field on a single farm⁷⁴.

While low crop yields and high agronomic yield gaps are linked to biophysical factors and management, it is the economic yield gap, shaped by input and output prices, that ultimately determines farmers’ investment decisions and returns on investment in fertilizer or organic input use. While profitability is a key driver for adoption, many other factors affect the uptake of appropriate fertilizer and soil health recommendations, including farmers’ production objectives and resource endowment, land tenure, and access to markets.

Regardless of the agro-ecological complexity, to enhance soil health while retaining economic viability, the combined use of mineral fertilizers with organic inputs and complementary good agronomic practices must be catalyzed through policy and other incentives to ensure a sustainable provision of food and other ecosystem services. The action plan, as laid out by the Fertilizer and Soil Health Summit of the African Union⁶ proposes critical measures to achieve increased and profitable crop yields and soil health.