Cacao is a strategic crop in many tropical countries, but achieving a balance between productivity, sustainability, and food security represents a complex challenge. This study presents a robust, multidimensional evaluation of five cacao production systems that integrates the energy-water-emissions-food (EWEF) nexus framework with life cycle assessment (LCA) and nutritional functional units to assess long-term environmental sustainability. Based on a 12-year dataset (2010-2021) from a long-term trial, the analysis spans from the establishment of the plots to their maturity. The systems include full-sun monocultures and agroforestry under organic and conventional management, and a successional agroforest, i. e., a highly diverse and dense system without external inputs. Monocultures, especially conventional ones, achieved higher cacao yields (between 1.77 and 2.22 times) but worse environmental outcomes (between -2.5% and -94%). In contrast, agroforestry systems produced 5.2 times more food (cacao and associated crops) per hectare, improving resource use efficiency in terms of energy (3.6 times more efficient) and using less water per kilogram of total food produced (-96%). Additionally, nutritional quality indicators reached higher values in relation to water and energy use in agroforestry systems (4 to 14 times higher), mainly in the successional one. Compared with monocultures, agroforestry systems had lower environmental impacts across most of the analyzed categories, during both the young (2010-2016) and the mature phase (2017-2021). Organic management did not systematically report lower environmental impacts for all categories and functional units analyzed. Our findings indicate that successional agroforests combine efficient resource use with high nutritional quality, and achieve similar cacao yields to those of conventional and organic agroforestry systems. Overall, this study supports the potential of diversified agroforestry systems-especially successional agroforestry-to advance food security and sustainability goals in cacao production.

Current pruning recommendations for clonal cacao (Theobroma cacao) lack scientific support regarding tree architectural development. This study investigates the relationship between tree architecture and production potential using data from a long-term trial in Bolivia. We measured architectural and production traits on 331 seventeen-year-old grafted trees across five cropping systems, ranging from conventional monocultures to input-free dynamic agroforestry, and four genotypes. We differentiated the productive “leafless zone” from the “leafy crown,” measuring branch orders, diameters, lengths in the former, and crown geometry traits in the latter. Results demonstrate that, while cropping systems and genotypes significantly influence vegetative biomass, the structural drivers of pod production remain consistent. Production was strongly related to the vigor (diameter) and number of low-order scaffold branches (orders 1 and 2) after accounting for cropping system type and genotype. Modelling confirmed that trees with complex scaffold structures and thicker 1st-order branches produced significantly more pods and beans than those with simple structures, with crown projected surface and foliage height further enhancing production potential. These findings challenge standard pruning recommendations focusing mainly on leafy crown management. We conclude that production potential is determined by the development of scaffold branches in combination with effective crown management. Consequently, pruning strategies should evolve from simple canopy control toward preserving vigorous scaffold branches and optimizing the tree’s physiological capacity for fruit production. To support this shift, we propose an architectural typology based on the branching structure of the leafless zone in grafted cacao trees to assess production potential and guide pruning decisions.

Diversification of farming systems at plot scale is increasingly seen as an essential option to meet agriculture sustainability challenges and support agroecological transition. Significant progress has been made in recent years on methods for designing diversified cropping systems based on annual crops. However, few methods have been tested for designing diversified cropping systems based on perennial crops, and in particular perennial crop-based agroforestry systems, despite their important dietary and economic roles in the world. In this study, we argue that the characteristics of perennial crops induce specific structural and functional changes that impact the design process. We identified four key concepts for the design of diversified perennial crop-based agroforestry systems: (i) the targeted productivity must be considered over the long term and at the plot level, and encompass not only the productive and non-productive stages of perennial crops but also of associated plants; (ii) these systems will eventually face shocks and crisis in their lifetime, so they should be designed with resilience and adaptability in mind; (iii) initial designs will necessarily evolve throughout the long lifespan of perennial crops, and so a combination of de novo design from scratch and step-by-step design approaches for regeneration or transformation will be required; and (iv) due to the high complexity of these systems and the difficulty of setting up trials, the design process should draw from a range of knowledges from academics and practitioners and hybrid them. These four key features show that perennial crops offer distinct opportunities for the design of sustainable diversified cropping systems, but also face specific challenges that require more transdisciplinary research.

Vertical canopy stratification generates strong microenvironmental gradients that shape plant physiological performance and resource-use strategies. Yet, the mechanistic pathways linking canopy structure, microenvironment, and trait-mediated responses remain poorly understood. In this study, we investigated how direct and indirect effects of vertical environmental heterogeneity, vegetation structure, and functional leaf traits regulates instantaneous water-use efficiency (WUE inst ), CO 2 assimilation rate ( A ) and transpiration rate (E) across four canopy strata in cacao agroforestry systems in Bolivia. A total of 546 leaves from 24 tree species were measured under experimental field conditions using wireless infrared gas exchange sensors. Leaf structural [Leaf area (LA), Specific leaf area (SLA)] and functional traits [Relative water content (RWC), Stomata density (SD)] were also quantified. We applied structural equation modelling to disentangle the mechanistic pathways regulating plant physiological responses across vertical strata. Our results reveal clear functional niche differentiation along the vertical canopy gradient. Upper-strata trees exhibited a resource-acquisitive strategy, characterized by increased CO 2 assimilation that outpaced transpiration resulting in a high WUE inst under high photosynthetically active radiation (PAR). In contrast, lower-strata trees are immersed in buffered microenvironmental conditions, with stable humidity and temperature. Under low PAR, these trees displayed a resource-conservative strategy, with reduced A and E and relatively constant WUEinst. While LA and RWC increased with canopy height, SLA and SD remained stable, indicating functional convergence of leaf traits supporting carbon gain and light tolerance in the upper strata. Overall, our findings demonstrate that canopy stratification drives contrasting water-carbon use strategies through, coordinated, trait-mediated responses to environmental gradients.

Cacao production is facing challenges of low productivity due to low soil fertility and climate change. Agroforestry and organic farming are potential sustainable and climate-resilient alternatives, but they are often associated with lower yields compared to monocultures and conventional farming. Despite their potential, empirical data on the long-term productivity of cacao cultivated in complex agroforestry systems and under organic management remains limited. Expanding this evidence base is essential to inform the development of agricultural practices and policies that advance environmental sustainability and food security. To fill this gap, we present 15 years (2008–2022) of data on cacao production and associated crops of a unique long-term trial comparing five cacao cropping systems in Bolivia: organically and conventionally managed monocultures, diverse agroforestry systems under organic and conventional management, and successional agroforestry systems without external inputs. We collected data on yields along with detailed information on the design and agronomic management from the beginning of the trial. All systems achieved competitive cacao yields in the mature phase. Organic and conventional systems had similar cacao yields, while agroforestry systems reached 56% of monoculture yields. Total system yields of the agroforestry systems were up to 6.9 times higher than monocultures. In the successional agroforestry, 22 crops were harvested, with short life cycle crops contributing to one-third of total production. This study shows that staple food crops and fruit trees as well as high-value crops (coffee, ginger, curcuma) can be successfully combined with cacao, and that agroforestry designs can be adapted over time by adding or eliminating crops to meet new goals or market opportunities. Extensive research has highlighted the positive contributions of agroforestry and organic farming to the delivery of ecosystem services. This study provides empirical evidence that it is possible to design and implement systems that reconcile environmental sustainability with productive performance.

Agroforestry systems are perceived as an effective approach to store carbon in agroecosystems by building tree biomass and raising soil organic carbon (SOC) stocks. This is especially evident in the tropics, where the cultivation of cash crops such as cacao in agroforestry systems is increasingly used. Among agroforestry systems, organic management, which avoids synthetic inputs for crop protection and fertilization, and the concept of successional agroforestry (SA), which aims to increase carbon storage by using high initial tree densities and intensive pruning without external inputs, have gained interest as alternatives to monocultures with less environmental impact. To assess the temporal development of carbon storage of differently managed agroforestry systems, we revisited a 14-year field experiment located in the Alto Beni Region of Bolivia to quantify biomass and SOC stocks in five distinct cocoa production systems. The field experiment includes SA as well as organic and conventional monocultures (OM and CM) and agroforestry systems (OA and CA). We found that all agroforestry systems increased carbon stocks in the biomass and the soil, especially in the particulate organic matter fraction. No significant effect of organic management practices was observed. After 14 years, the highest biomass carbon was observed in the SA system and topsoil SOC stocks increased significantly in SA and CA. Our findings emphasize the potential to enhance carbon accumulation in agroforestry systems with high initial tree density and rigorous pruning, even without additional fertilizer or synthetic plant protection inputs.

Previous studies suggest that the density and diversity of associated trees stabilises cocoa agroforestry systems (cAFS) production in the long term. However, the impact of cocoa rejuvenation practices on this stabilisation process remains unclear. This study investigated how farmers’ practices, particularly cocoa stand management, influence long-term cocoa yields, considering the production area’s history and current management. Cocoa plantations aged from 7 to over 60 years were selected in two areas: an old cocoa production area (Bakoa) and a cocoa pioneer front in the forested land (Talba) in Cameroon’s Centre region. Measurements included cocoa and associated tree diameters, shade tree cover and accessible cocoa yield. We also identified architectural types of cocoa trees representing various development stages. Results showed no effect of plantation age on associated tree parameters at either site. In Talba, cocoa tree density and accessible yield decreased with plantation age. This decline was associated with an increase in senescent cocoa trees and few regenerated multi-stem trees over time. This indicates poor rejuvenation of existing plantations in Talba, where forest reserves are available to create new ones. Conversely, in Bakoa, an area with no available forest reserves, rejuvenation of existing plantations, evidenced by the increasing density of multi-stem cocoa trees and new seedlings over time, was associated with stable yields in the long term. Providing farmers with training and incentives to rejuvenate old cocoa stands should be part of an overall strategy to increase production, improve farmers' incomes, protect forests and comply with the commercial rules of consumer countries.

Cacao in agroforestry systems can benefit from improved nutrient cycling, since shade trees could transfer nutrients through litterfall and pruning residues. Additionally, shade trees could affect decomposition processes of these residues. Studies on nutrient release from pruning residues and on the effect of shade trees on decomposition processes are scarce. We aimed to determine how cropping systems (agroforestry, monocultures) affect the decomposition of pruning residues and quantify nutrient release by these residues. Litterbags with two mesh sizes (0.1 and 2 mm), containing leaf mixtures, were placed under five cropping systems (conventional and organic monocultures, conventional and organic agroforestry, successional agroforestry) in a long-term trial in Alto Beni, Bolivia. Carbon, phosphorus, nitrogen, potassium, cellulose, and lignin were measured in fresh leaves and litterbags at 4, 8, and 12 months after laying. Nitrogen was higher under conventional agroforestry than under the other systems at 4 months after laying. Phosphorus was higher under agroforestry than under monocultures at 8 months after laying. Litterbags with 2 mm mesh size contained lower amounts of carbon and nutrients at 4 months after laying than litterbags with 0.1 mm mesh size. Release of nutrients from pruning residues was estimated. The effect of shade trees on decomposition processes was limited to changes in nitrogen and phosphorus contents in litterbags. Larger mesh litterbags likely favoured microbial colonization and nutrient transfer through leaching. The nutrient input from pruning residues could meet the potassium and nitrogen demands of cacao but is insufficient for phosphorus.

Cacao Nacional Boliviano (CNB), the native cacao germplasm group in the Bolivian Amazon region, possesses distinct morphological, genetic, phenological and organoleptic characteristics. It is highly sought after in the global fine-flavour chocolate market, and has importance for conservation of cacao genetic diversity. However, CNB cultivation is minor and yield levels low. Aiming to develop profitable CNB cultivation practices adapted to its distinct phenology and genetic basis, we compared parameters on CNB flowering, yields, biomass and harvesting effort in three agricultural systems with and without pruning and varying management intensity: agroforestry with pruning (AF), underplanted secondary forest with (SFwP), and without pruning (SFnP). Repeated measures linear mixed models using Restricted Maximum Likelihood were applied for statistical analysis. While AF and SFnP yielded 127.6 ± 21.4 kg ha−1 and 212.4 ± 22.2 kg ha−1 on average, reaching > 300 kg ha−1 after 11 and 9 years, respectively, SFwP yields remained significantly lower than SFnP, with an average of 58.0 ± 9.5 kg ha−1 and maximum yield of 122.6 ± 36.5 kg ha−1. This study demonstrates there may be a trade-off between pruning intensity and yield in AF and SFwP, as observed on young CNB trees’ yield developments and yield increases during two years without pruning in mature trees. Based on preliminary results and CNB-specific phenology, we suggest a moderate pruning intervention early in July (allowing recovery time before start of flowering season in August), along with the need for validation of our findings and further investigation into management practices tailored to CNB.

An estimated 3.4 million hectares of cocoa and 9.7 million hectares of coffee are cultivated, globally, under shade trees, i.e. in agroforestry systems. Shade canopies are characterized in terms of tree density (N, trees ha −1 ), tree basal area (G, m 2 ha −1 ) and percent canopy cover (%Cov). N, G and %Cov are named shade canopy density variables (SCDV). The use of these SCDV has two important limitations: (1) different combinations of values of the three SCDV variables generate very different shade tree stands (hence very different shading levels), and (2) Additional factors modify shading under shade canopies with constant SCDV values. This article uses the software ShadeMotion ( www.shademotion.net ) to show how 24 different, simple, even-sized, mono-layered, Cordia alliodora shade canopies with constant N, G and %Cov display significantly different shade levels and temporal patterns of shading depending on tree stem and crown diameter ratios, tree height, spatial planting configurations (square, random and alleys) and leaf fall patterns. A minimum set of variables capable of providing a more accurate description of the shading characteristics of a cocoa or coffee shade canopy is proposed. Our findings can shed light on the current debate on the pros and cons of the definitions of cocoa agroforestry used by chocolate and certification companies, governments, non-governmental organizations, and donors, especially in West and Central Africa. In this article, emphasis is given to cocoa, but the analysis, results and conclusions are equally applicable to coffee agroforestry systems.

Organic agriculture and agroforestry are two agroecological approaches that have been proposed to address the many negative externalities of intensive agriculture.However, their comparative efficiency in promoting sustainability when combined is unclear, as they are typically studied in isolation. To address this we conducted a systematic review of comparative studies addressing organic versus conventional agroforestry and their monocultural counterparts. We conducted a content analysis resulting in an impact matrix utilizing seven broad categories encompassing ecological, socioeconomic and environmental aspects of sustainability. By analyzing these impact categories separately, it is possible to highlight the distinct advantages and issues in organic agroforestry systems over alternative systems, as well as the potential for multifunctionality. Research in organic agroforestry is strongly biased towards South American Cocoa and Coffee, but, in spite of this, evidence thus far suggests that organic agroforestry has many advantages across all categories studied and few drawbacks; the main one being yield of single crops. A large number of comparisons yielded neutral outcomes, suggesting that there are a limited set of trade-offs associated with organic agroforestry, many of which may be attenuated by altering management and market conditions. We conclude by highlighting key research areas in organic agroforestry which need to be addressed including: the differing ways of quantifying yield, staple crops and expanding the geographic range of studies.

A network of agronomists, researchers, and practitioners associated with cacao farming provided open access to their independent field trials across Latin America and the Caribbean (LAC). A centralized dataset was assembled using qualitative and quantitative data from 25 experimental field trials (hereafter referred to as “CacaoFIT”) spanning several LAC agroecosystems. This dataset was used to document the main traits and agroclimatic attributes of the cacao cultivation model being tested within the CacaoFIT network. By synthesizing data from an entire network of cacao trials, this study aimed to highlight specific design features and management practices that may contribute to better cacao farming sustainability. The CacaoFIT network comprises 200 ha of field trials testing over 150 cacao genotypes and set up under different shade canopy design, management, and research goals. Small-sized trials were common across Mesoamerica, whereas medium to large-size trials were distinct to South America. Cacao trials were 15 years old (on average) and ranged from 3 to 25 years of establishment. Most cacao trials were managed conventionally (i.e., 55%), while 20% were under organic practices, and the remaining 25% presented both conventional and organic management approaches. Most field trials (ca. 60%) planted an average of 10 international clones or national cultivars at high (1,230–1,500 plants ha−1) and medium density (833–1,111 plants ha−1). Mixed shade canopies were the dominant agroforestry model, while timber vs. leguminous shade canopies were also common. The diversity and depth of research domains examined across the CacaoFIT network varied widely. Agronomy and agroforestry topics dominated the research agenda across all trials, followed by environmental services domains. Cacao physiology and financial performance were researched to a lesser extent within the network. Five featured field trials from CacaoFIT offered technical guidelines to inform cacao farming within similar contexts. This collaborative work is a scaffold to encourage public–private partnerships, capacity building, and data sharing amongst cacao researchers across the tropics.

Agroforestry systems (AFS) are gaining major interest in cocoa producing countries. However, the analysis of competition effects in multispecific complex agricultural systems remains a key lock-in preventing both (i) a thorough understanding of resource partitioning, and (ii) the co-design of spatial patterns and management of cocoa AFS (cAFS) minimising competition. We sampled 48 cAFS 1000 m2 plots in Central Cameroon and used an individual based approach to analyse the effect neighbouring tree community composition had on coca tree pod production. First, we determined the distance up to which each neighbouring tree influenced the production of a given cocoa tree and checked for the significance and the magnitude of this effect. We then explored the trade-offs between cocoa tree production and the abundance of associated trees. Our results first underline a significant intra-specific competition within the cocoa tree populations studied, which would need to be thinned down to 800-1100 individuals per ha. We also found a general negative effect of associated trees kept or grown for their fruits (e.g. mango, traditional species, citrus trees). Those effects were stronger when the trees were positioned between 6 and 11 m away from the cocoa trees. These trees clearly impaired pod production, especially for small diameter cocoa trees. Palm trees, however, had a positive effect on cocoa pod production, with suitable densities modelled ranging from 0 to 240 individuals per ha. Finally, we found both positive and negative effects of associated forest trees on cocoa pod production, which varied with the distance separating them from the cocoa trees and the size of the cocoa trees. Overall, our analyses show that cocoa production in complex cAFS is influenced both by intra-and inter-specific interactions while it remains difficult to distinguish between potential intertwined effects and resource limitations.

Duru et al. (Agron Sustain Dev 35:1259-1281, 2015) highlighted a missing tool for studying and improving the performance of cropping systems in the transition to highly diversified agriculture. In response, this paper proposes a concept for designing, modeling, monitoring, and auditing desired ecosystem services, in intercropping and agroforestry systems. We have labelled this concept ESSU (Ecosystem Services functional Spatial Unit). It delimits the smallest spatial unit encompassing all the interacting species and other functional components (e.g., crops, trees, livestock, spontaneous vegetation, semi-natural habitats such as hedges, ditches, forest patches, and animals) that together provide a specified set of ecosystem services. The novel ESSU concept allows representation of an entire diversified agroecosystem by the repetition of the spatial unit that provides the same sets of targeted ecosystem services as the agroecosystem it represents. It can then be used for various activities, such as the (i) design of more efficient agroecological systems according to the targeted ecosystem services; (ii) rapid audit of farming practices for biodiversity/resilience across large tracts of farmland as part of achieving Sustainable Development Goal 2 targets of sustainable food systems; and (iii) modeling such diversified agroecosystems using a motif adapted to represent the targeted ecosystem services and the species spacing design. We demonstrate that the ESSU concept is highly flexible and applicable to a wide range of diversified agroecosystems, like arable intercropping, crop-tree intercropping, tree-tree agroforestry, and agro-pastoralism. We also show its relevance and suitability for representing temporal changes over 1 year, across several years, and over decades, indicating its generalizability and flexibility. We argue that ESSU could open new theoretical and practical research avenues for the study of diversified agroecosystems. Considered with all the knowledge available on practices, biodiversity, and ecosystem services, ESSU might provide a learning-support tool to fill the knowledge gap about relationships among practices, biodiversity, and associated ecosystem services.

Description. La culture du cacaoyer en Guyane tend à se (re)développer et nécessite aujourd’hui des références technico-économiques adaptées. Objectifs. Réalisation d’une évaluation de la viabilité technico-économique de la cacaoculture et de la transformation de ses produits.Méthode. Modélisation des produits et des couts de l’exploitation d’un hectare selon deux modes de gestion différents et trois types de produits finis. Résultats. Le système de culture installé après défriche partielle d’une forêt demande moins de main-d’œuvre et de trésorerie que le système s’appuyant sur l’association au bananier qui, lui, permet un retour sur investissement avant maturité de la production cacaoyère. Pour être viables, les systèmes de culture ayant pour seule finalité la production de fèves fermentées et séchées nécessiteraient des rendements de l’ordre d’1 tonne·ha-1 pour des prix de vente planchers souvent supérieurs à 10 000 €·t-1, ce qui interroge fortement sur leur faisabilité. Pour la production transformée en bâton de cacao ou en chocolat, la viabilité économique serait plus facilement acquise, mais demanderait un investissement et une technicité qui ne sont pas forcément accessibles aujourd’hui à toutes les exploitations qui s’engagent dans la cacaoculture.Conclusions. Nos résultats soulignent la nécessité de (re)considérer dans les régions ultramarines françaises : (i) la nature et l’usage des espèces associées aux cacaoyers ; (ii) le statut de culture principale ou non du cacaoyer dans les exploitations agricoles ; (iii) la diversification des produits du cacaoyer permettant de nourrir plusieurs marchés ; (iv) l’organisation collective de la filière naissante pour soutenir la production et la transformation ; (v) l’accompagnement technique et financier (ex : aides PAC) des agriculteurs se lançant en cacaoculture.

Description of the subject. The cultivation of cocoa trees in French Guiana is currently experiencing a renewed level of interest and activity. A suitable framework is therefore needed to study the viability of this emerging area of cultivation. Objectives. To carry out an assessment of the technico-economic viability of cocoa cultivation and the processing of its products. Method. The costs and benefits of two different cropping systems and three types of end products were modeled. Results. The cropping system installed after partial forest clearing requires less labor and cash flow than the system based on association with banana, which allows a return on investment before the maturity of cocoa production. In order to be viable, cropping systems whose sole purpose is the production of fermented and dried beans would require yields of approximately I ton.ha(-1) and floor prices often above (sic) 10,000.t(-1). Such a result questions the feasibility of these systems. If beans were processed into cocoa sticks or chocolate, economic viability would be more easily acquired, but this would require investments that are not necessarily compatible with the cash flow of farms currently involved in cocoa fanning. Conclusions. These results highlight the need to (re)consider in French overseas regions: (i) the nature and use of species associated with cocoa trees; (ii) the status of the cocoa tree as the main crop (or not) on farms; (iii) the diversification of cocoa products making it possible to feed several markets; (iv) the collective organization of the emerging sector to support production and processing; (v) technical and financial support (e.g.: CAP aid) for fanners embarking on cocoa farming.

Plant diversification through crop rotation or agroforestry is a promising way to improve sustainability of agroecosystems. Nonetheless, criteria to select the most suitable plant communities for agroecosystems diversification facing contrasting environmental constraints need to be refined. Here, we compared the impacts of 24 different plant communities on soil fertility across six tropical agroecosystems: either on highly weathered Ferralsols, with strong P limitation, or on partially weathered soils derived from volcanic material, with major N limitation. In each agroecosystem, we tested several plant communities for diversification, as compared to a matching low diversity management for their cropping system. Plant residue restitution, N, P and lignin contents were measured for each plant community. In parallel, the soil under each community was analyzed for organic C and N, inorganic N, Olsen P, soil pH and nematode community composition. Soil potential fertility was assessed with plant bioassays under greenhouse controlled climatic conditions. Overall, plant diversification had a positive effect on soil fertility across all sites, with contrasting effects depending on soil type and legumes presence in the community. Communities with legumes improved soil fertility indicators of volcanic soils, which was demonstrated through significantly higher plant biomass production in the bioassays (+18%) and soil inorganic N (+26%) compared to the low diversity management. Contrastingly, communities without legumes were the most beneficial in Ferralsols, with increases in plant biomass production in the bioassays (+39%), soil Olsen P (+46%), soil C (+26%), and pH (+5%). Piecewise structural equation models with Shipley's test revealed that plant diversification impacts on volcanic soil fertility were related to soil N availability, driven by litter N. Meanwhile, Ferralsols fertility was related to soil P availability, driven by litter P. These findings underline the importance of multifactorial and multi-sites experiments to inform trait-based frameworks used in designing optimal plant diversification in agroecological systems.

Manipulating plant functional diversity to improve agroecosystem multifunctionality is a central challenge of agricultural systems world‐wide. In cocoa agroforestry systems (cAFS) shade trees are used to supply many services to farmers,yet their impact on soil functioning and cocoa yields is likely to vary substantially among tree species. Here we compared the impact of five shade tree species (Canarium schweinfurthii (Canarium), Dacryodes edulis (Safou), Milicia excelsa (Iroko), Ceiba pentandra (Kapok tree), Albizia adianthifolia (Albizia)) and unshaded conditions on the functioning of poor sandy savanna soils within eight cocoa farms in Central Cameroon. We assessed the effects of plant functional traits, leaf litterfall and fine root biomass on a range of soil functions and on cocoa yield. Shade trees generally improved soil pH,and Olsen P content, biomass production of bioassays and soil total C and N content, while leaving cocoa yields unchanged. However, these effects varied largely among species. Improvements of soil functions were low under the two fruit trees (Canarium and Dacryodes), medium under the legume tree Albizia and high under the two timber trees (Milicia and Ceiba). Low litter recalcitrance was most strongly associated with increases in soil fertility indicators such as N and P availability, whereas soil C and N content increased with litter Ca restitution. Synthesis and applications. We demonstrate that cocoa agroforest multifunctionality is substantially influenced by the functional traits of shade tree species. Shade tree species with the most dissimilar traits to cocoa (cocoa showing the lowest leaf litter quality) showed the largest improvement of soil functions. Therefore, selection of shade trees based on their functional traits appears as a promising practice to adequately manage soil functioning. In order to fully assess the beneficial role of shade trees in these agroecosystems. Future research will need to extend this approach to other below‐ground traits and other aspects of multifunctionality such as long‐term cocoa health and yield.

Combining mixed trees with cocoa trees in so-called cocoa agroforestry systems is emerging as a viable option for regenerating cocoa cropping in Africa. Pure cocoa crop stands currently prevail in these areas but are running out of steam. Technical solutions are therefore urgently needed to stabilize cocoa-growing areas, reduce pressure on forests and adapt to climate change. A balance can be struck in stands by combining cocoa trees with diverse trees chosen by farmers for their different uses, while maintaining a suitable and sustainable cocoa yield. Ecological services such as carbon storage and crop protection are co-benefits of this balance. Recent research in Cameroon highlights that cocoa agroforestry stands can be managed using a straightforward indicator—measurement of the basal area of cocoa trees and associated trees. This indicator could be adopted for sustainable cocoa production certification purposes, while the observed convergence between local know-how and scientific results could facilitate joint drawing up of technical recommendations.