---

Gabésongon Kone¹ , Abo Kouabenan¹ , Mariam Bognan Coulibaly²

¹Institut National Polytechnique Félix Houphouët-Boigny (INP-HB), Yamoussoukro, Côte d'Ivoire.

²Université Jean Lorougnon Guédé (UJLoG), Daloa, Côte d'Ivoire.

Corresponding Author Email: bognanmari100@gmail.com

DOI : https://doi.org/10.51470/JPB.2026.5.1.41

Abstract

Keywords

Download this article as:

---

Introduction

The cashew tree (Anacardium occidentale L.), a perennial species belonging to the Anacardiaceae family, is native to northeastern Brazil. Widely recognized for the significant economic value of its fruit—the cashew nut—the species was introduced to Africa and Asia by the Portuguese [1]. Although initially deployed to combat soil erosion, it has since evolved into a major cash crop across West Africa [2]. This transition is driven by its high market value, the increasing organization of the value chain, and its vital socioeconomic role in rural communities.

Currently cultivated in nearly all tropical regions, cashew nuts hold a strategic position in Côte d’Ivoire, which emerged as the world’s leading producer of raw cashew nuts (RCN) in 2019 [3]. By 2023, Ivorian production reached an estimated 1,225,935 tons, accounting for approximately 40% of global output [4].

The cashew tree plays a fundamental role in the socioeconomic landscape of rural populations in Côte d’Ivoire. Much like cocoa, its cultivation enables the northern and northeastern regions to bolster financial resources and improve access to essential education and healthcare services. Cashew farming is predominantly characterized by smallholder family farms, with plots typically ranging from 0.5 to 3 hectares. These systems frequently employ intercropping with food crops, a strategy that not only optimizes land utilization but also reduces maintenance costs; specifically, the expanding canopy of the cashew trees naturally suppresses weed growth.

Revenue generated from the marketing of cashew nuts is further utilized to fund social ceremonies (such as weddings, funerals, and rituals), acquire consumer goods (including motorcycles and appliances), and invest in housing construction [5]. Overall, the cashew sector supports approximately 2.5 million Ivorians and contributes 7% to the national Gross Domestic Product (GDP) [5].

Despite its socioeconomic importance, the average yield per hectare remains relatively low, estimated at 620 kg/ha, which is significantly below the optimal potential of 1,200 kg/ha for raw cashew nuts (RCN). This yield gap is attributed to several factors, including insufficient agricultural investment, limited knowledge regarding input application, and a lack of proficiency in essential cultural practices, such as pruning. Previous studies have highlighted these challenges, emphasizing that specific agricultural behaviors and technical constraints directly contribute to the low productivity of orchards across the region [7, 8, 9].

Notwithstanding the alarming data on low yields, comprehensive knowledge of tree cropping systems in West Africa remains limited and fragmented. Whether at the micro-scale (individual tree) or the macro-scale (plot or production basin), available data are frequently imprecise, incomplete, or absent

In Côte d’Ivoire, while cashew nuts occupy a strategic position in rural economies—serving as an essential income source for producers in the north and northeast—the crop’s potential remains largely under-exploited. This underperformance is compounded by high yield variability, which remains poorly documented at both the tree and orchard levels. A deeper understanding of this variability is therefore critical to identifying primary limiting factors and developing tailored agronomic interventions.

In this context, the present study aims to analyze cashew nut yield variability at two distinct levels: the individual tree and the orchard. Specifically, the objectives are to:

• Evaluate yield evolution as a function of tree age and planting density.
• Analyze the influence of tree morphological structure (height, trunk diameter, and canopy) on productivity across different densities and age groups.

Research Hypotheses

H1: Tree productivity varies significantly based on age and planting density.

H2: Tree morphology (height, trunk diameter, and canopy diameter) is significantly influenced by both planting density and tree age.

METHODS

Materials And Methods

This section details the resources and procedures implemented to conduct the study. It is organized into two primary components: Materials, which include a description of the study site, plant material, and measurement tools; and Methods, which provide a detailed account of the data collection protocols and statistical analyses.

Study Site

The study was conducted across 24 orchards in the Poro region (8°26’–10°27′ N, 8°26’–10°27′ W). Covering an area of 13,400 km², the Poro region is organized into four departments: Dikodougou, Korhogo (the regional capital), M’Bengué, and Sinématiali. The regional climate is Sudanese, characterized by a distinct dry season from November to April and a rainy season from May to October [10]. Annual rainfall typically ranges from 1,000 to 1,400 mm, with average monthly temperatures between 26.93°C and 27.02°C, peaking at approximately 36°C in March [11]. The vegetation is diverse, featuring Sudanese savanna in the northern reaches and sub-Sudanese savanna in the south. The topography is varied, with elevations such as Mount Korhogo exceeding 500 m. Soils are predominantly ferralitic, ferruginous, and hydromorphic, characterized by high permeability and porosity. The local economy is primarily driven by livestock and agriculture, including major cash crops (cotton and cashew) as well as food crops (yam, rice, and sorghum).

 Materials

This subsection details the equipment used, ranging from field data collection (morphological and yield parameters) to laboratory-based statistical processing.

Plant Material

The plant material consisted of cashew trees (Anacardium occidentale L.) located within 24 orchards delineated into study plots in the Poro region.

Technical Equipment

Cashew tree dimensions were measured using a decameter, while a graduated wooden pole was used for height measurements. Plot boundaries (2,500 m²) were established using ropes, and the seven selected sample trees per plot were marked with spray paint. For nut collection and measurement, harvest bags and a precision scale were used. Data (nut mass and quantity per tree and per plot) were initially recorded in field notebooks and subsequently digitized using Open Data Kit (ODK).

Methods

Orchard Selection

To analyze cashew yield variability at both the tree and plot scales, four age categories were selected: 5, 10, 20, and 30 years. These categories were chosen to better understand yield fluctuations and tree physiological behavior over time.

Data Collection Design

A preliminary survey was conducted to assess orchard age and density. Consequently, the selection focused on these two factors. For each of the four age categories (5, 10, 20, and 30 years), three planting density classes were defined:

• Low density: ~75 trees/ha
• Medium density: ~150 trees/ha
• High density: ~300 trees/ha

Plot Establishment: Based on these criteria, 18 orchards were selected (2 orchards per density class × 3 density classes × 4 age categories — Note: Please verify if the total is 18 or 24 based on your math). Within each orchard, a 50 m × 50 m (2,500 m²) plot was established, and seven specific trees were monitored within each plot.

Morphometric Measurements

Morphometric characteristics were evaluated for all seven numbered trees per plot. Measurements included tree height, trunk diameter (DBH), and canopy diameter. Height was determined using a graduated wooden pole, while trunk and canopy diameters were measured with a measuring tape and decimeter, respectively.

Data Analysis Methodology

A multi-step approach was employed to analyze yield variability. First, descriptive statistics were performed to explore data distribution and identify outliers. Second, graphical analyses were used to visualize variable trends relative to planting density and tree age.

One-way Analysis of Variance (ANOVA) was performed to determine the effect of categorical factors on quantitative variables. To ensure the reliability of the results, ANOVA assumptions— specifically normality of residuals, homogeneity of variances, and independence of observations—were rigorously tested. Finally, an Ordinary Least Squares (OLS) multiple linear regression model was used to evaluate the impact of morphological variables (height, canopy diameter, and trunk diameter) on yield per tree. Model validation was based on the fulfillment of normality, homogeneity, and independence assumptions [12]. Data processing and statistical analyses were conducted using Microsoft Excel and RStudio.

 RESULTS

This section presents the detailed findings from the statistical analyses and field observations conducted in this study. It begins with a descriptive and exploratory analysis of yield variability and morphological parameters, followed by hypothesis testing (ANOVA) and the results of the multiple linear regression modeling regarding the influence of morphological traits on yield.

 Descriptive Analysis

Descriptive and exploratory data analysis allowed for the characterization of the statistical distribution of yield variables and morphological parameters (height, DBH, and canopy diameter). Table 1: Descriptive statistics of morphological and productive characteristics of cashew trees

Table 1 presents the descriptive statistics for the primary morphological characteristics (height, diameter at breast height, and canopy diameter) and yield per tree.

Regarding morphological parameters, tree height ranged from 3.93 m to 7.34 m, with a mean of 5.93 m. The proximity between the mean and the median indicates relatively homogeneous growth; however, the gap between the extreme values suggests differences in development likely related to tree age or planting density. Furthermore, the DBH, ranging from 13.51 to 46.04 cm (mean = 22.84 cm), and the canopy diameter, varying from 4.77 to 11.29 m (mean = 7.71 m), confirm this trend, revealing a majority of medium-sized trees.

In contrast, yield results indicate high inter-tree variability, fluctuating from 0.95 to 39.33 kg/tree, with a mean of 11.20 kg/tree. Since the median (9.09 kg/tree) is lower than the mean, the distribution is positively skewed, heavily influenced by a few highly productive trees. Consequently, this heterogeneity suggests that productivity depends not only on tree morphology but also on other factors, including genetic variability, micro-environmental conditions, and cultural practices.

Thus, at the individual tree level, productivity appears highly unequal, while at the orchard scale, this intra-plot variability directly influences overall performance and production consistency.

 Univariate Analysis of Variance for Morphological Variables

Table 4 summarizes the results of the one-way ANOVA conducted to evaluate the combined effects of tree age and planting density on morphological characteristics, specifically Diameter at Breast Height (DBH), canopy diameter (crown spread), and tree height.

Table 4: Univariate ANOVA for morphological variables.

The results indicate that planting density and age have a significant effect on the trunk diameter growth of cashew trees (F = 3.983; p = 0.0126). Regarding the mean canopy diameter, a highly significant effect was observed (F = 5.757; p = 0.00268), reflecting the sensitivity of crown expansion to variations in density and age. Mean tree height was also significantly influenced by these factors (p < 0.05).

ANOVA Diagnostic Tests on Residuals

To ensure the validity of the ANOVA, the underlying assumptions were tested and are summarized in Table 5.

The results show that the residuals for DBH, canopy, and height yielded p-values greater than

0.05 for the Shapiro-Wilk, Bartlett, and Durbin-Watson tests. These findings confirm the normality, homogeneity of variances, and independence of the residuals, thereby validating the use of ANOVA for this analysis.

Multiple Linear Regression: Effects of Morphological Variables on Yield per Tree

Table 6 presents the results of the multiple linear regression model using the Ordinary Least Squares (OLS) method. This model was developed to assess the influence of height, canopy diameter, and DBH on yield per tree (Log-transformed).

Table 6: Effects of morphological variables on yield per tree.

The analysis reveals that DBH and canopy diameter are the primary morphological predictors of yield per tree, showing significant positive effects (p = 0.017 and p ≈ 0.054, respectively). This suggests that more developed radial growth and horizontal architecture favor individual tree production. In contrast, tree height did not have a significant effect on yield (p = 0.221), although a negative trend was observed. Overall, these results indicate that radial vigor (DBH) and horizontal spread (canopy) contribute more to productivity than vertical growth.

DISCUSSION 

This section interprets the results in light of the initial research hypotheses and existing scientific literature. Graphical and statistical analyses demonstrated that the orchard “type” (defined by age and planting density) significantly influenced yield (following log-transformation) at the 10% significance level. The fundamental assumptions for the ANOVA model, normality of residuals, homogeneity of variances, and the absence of autocorrelation, were validated through their respective diagnostic tests. These findings provide a robust basis to conclude that the various plot types studied exert distinct impacts on yield.

Influence of Density and Age on DBH 

The results indicate that trunk diameter (Diameter at Breast Height, DBH) is significantly influenced by planting density. While high-density configurations may favor vegetative growth in the short term, inter-tree competition for water, nutrients, and solar radiation becomes the primary limiting factor for radial development over the long term. Our observations corroborate the findings of [11], who demonstrated that excessive density triggers tree elongation (etiolation) as individuals compete for light at the expense of secondary (radial) growth.

Conversely, low-density arrangements, such as 75 trees/ha, facilitate superior resource acquisition and utilization by individual trees. This reduced competition fosters accelerated growth rates and results in significantly larger trunk diameters over time. This is in agreement with [12], who argued that wider spacing promotes lateral trunk expansion through optimized allometric growth. In summary, although high-density planting may offer initial advantages in terms of land-use efficiency, adequate spacing is critical for ensuring sustained physiological health and achieving optimal trunk dimensions in cashew orchards.

Canopy Development and Planting Density 

Tree age is a determining factor for canopy architecture, as the mean canopy diameter increases over time regardless of planting density. However, this growth is non-uniform; it becomes significantly more pronounced after 10 years, particularly in low-density plots. These results align with the observations of [12], who also highlighted the critical impact of density on canopy geometry and spatial distribution. 

Consequently, to optimize long-term canopy expansion, low planting densities are recommended. This conclusion is supported by [13], who stated that the optimal density for maximizing crown development is approximately 100 trees/ha. Beyond this threshold, inter-tree competition for light and space intensifies with age, eventually leading to canopy overlap and reduced individual tree vigor.

Vertical Growth and Resource Competition 

Cashew tree height is determined by both planting density and tree age. While high-density configurations may initially induce an appearance of rapid, vigorous vertical growth, an attribute often favored by farmers, our results demonstrate that such density levels are unsustainable for individual tree development over the long term. This initial height gain is often a physiological response to shading, where trees prioritize apical growth to access light.

However, as the orchard matures, below-ground competition for limited soil resources eventually restricts further development. Our observations align with the findings of [14], who showed that while high density stimulates height growth in juvenile trees, a threshold effect emerges over time, diminishing this advantage. In mature cashew orchards with established canopies, subterranean resources, pecifically soil moisture and nutrient availability become the critical limiting factors for sustained growth and tree longevity.

Morphological Determinants of Yield 

The multiple linear regression analysis evaluated the influence of tree height, canopy diameter, and DBH on individual tree yield. The results demonstrate that both DBH and canopy diameter exert a significant positive effect on yield (p=0.017 and p=0.054), respectively), while height had no significant effect (p=0.221).These findings suggest that radial vigor and horizontal architecture are stronger contributors to productivity than vertical growth.

This trend is consistent with the findings of [15], who identified trunk diameter and canopy spread as primary indicators of cashew productivity, largely due to their direct relationship with photosynthetic leaf area and the structural capacity to support fruit load. In contrast, the lack of significant correlation with tree height corroborates the conclusions of [16], which suggest that in perennial fruit crops, vertical elongation does not necessarily translate into higher yields and may even indicate resource allocation toward vegetative maintenance rather than reproductive output.

The multiple linear regression analysis evaluated the effect of height, canopy diameter, and DBH on yield per tree. The results show that DBH and canopy diameter exert a positive effect on yield (p=0.017 and p=0.054, respectively), while height had no significant effect (p=0.221). These observations suggest that radial vigor and horizontal architecture contribute more to productivity than vertical growth. This trend is consistent with that of, who showed that trunk diameter and canopy spread are key indicators of fruit tree productivity due to their relationship with photosynthetic leaf area and structural support for fruit. In contrast, tree height showed no significant correlation, corroborating the conclusions that vertical growth does not always correlate with higher production in woody species.

Conclusion 

The multiple linear regression and one-way Analysis of Variance (ANOVA) conducted in this study identified the primary factors influencing cashew yield at both the individual tree and orchard scales. Our findings confirm that Diameter at Breast Height (DBH) and canopy diameter are the morphological traits most directly linked to individual productivity. Specifically, DBH exerted a positive and significant effect on yield (p=0.017), while the canopy diameter also showed a positive influence, albeit with marginal significance (p 0.054). However, the results also revealed substantial yield variability among trees with similar morphologies, suggesting the influence of additional underlying factors. At the orchard level, statistical evidence demonstrated that planting density is a critical determinant of yield. Trees grown at lower densities exhibited superior long-term productivity, as wider spacing facilitates optimal trunk and canopy development by minimizing inter-tree competition for resources. Although high planting densities may initially stimulate vertical growth, they ultimately constrain individual development as competition for water and nutrients intensifies. Consequently, maintaining adequate spacing is essential for maximizing sustainable cashew production. 

Future Perspectives

To build upon the findings of this study, we propose integrating additional variables into future research. Subsequent studies should focus on the following key areas:

• Genotypic Analysis: Investigating the influence of different cashew genotypes to determine the extent to which genetic variation impacts yield and adaptability.
• Edaphic Factors: Evaluating the role of soil properties including soil type, pH, and nutrient fertility in driving tree growth and overall productivity.
• Climatic Variables: Modeling the effects of rainfall patterns, temperature fluctuations, and humidity levels on morphological development and yield stability.

 REFERENCES

[1]AZONKPIN, A. (2025). Performance des arbres-mères d’anacardier (Anacardium occidentale) pour une sélection orientée vers la production des plants greffés au Bénin. https://doi.org/10.5281/ZENODO.15540139  

 [2] Lautié, E., Dornier, M., De Souza Filho, M., & Reynes, M. (2001). Les produits de l’anacardier : caractéristiques, voies de valorisation et marchés [Cashew products: characteristics, recovery pathways and markets]. Fruits, 56(4), 235–248. https://doi.org/10.1051/fruits:2001126

[3] Maximin Lucien DA, & Da, M. L. (2024). Contribution de la production des noix de cajou à la croissance économique de la Côte d’Ivoire. Revue d’Economie Théorique et Appliquée, 14(1), 21‑40. https://doi.org/10.62519/reta.v14n1a2

[4] CICG. (2024). ANACARDE : La Côte d’Ivoire représente 40% de l’offre mondiale pour une production de 1 225 935 tonnes en 2023 [Cashew: Ivory Coast represents 40% of global supply]. Gouv.ci. http://www.gouv.ci/_actualite-article.php?recordid=16961 Dicostat. (n.d.). [Dictionary of Statistics].

[5] Yabi, I., Yabi Biaou, F., & Dadegnon, S. (2013). Diversité des espèces végétales au sein des agro-forêts à base d’anacardier à Savalou au Benin [Plant diversity within cashew-based agroforests in Benin]. International Journal of Biological and Chemical Sciences, 7(2), 696–

706. https://doi.org/10.4314/ijbcs.v7i2.24

[6] ANADER. (2025). [National Cotton and Cashew Producer Days]. www.anader.ci

Bobbitt, Z. (2020, December 31). The Breusch-Pagan Test: Definition & Example. Statology. https://www.statology.org/breusch-pagan-test/

[7] Monteiro, F., Catarino, L., Batista, D., Indjai, B., Duarte, M., & Romeiras, M. (2017). Cashew as a high agricultural commodity in West Africa: Insights towards sustainable production in Guinea-Bissau. Sustainability, 9(9), 1666. https://doi.org/10.3390/su9091666

[8] Akohoue, F., Segnon, A. C., & Achigan-Dako, E. G. (2018). Diversity in smallholder citrus orchards and cultivation bottlenecks: Research avenues for improved production in Benin, West

Africa. Experimental     Agriculture,      54(5),   641–654. https://doi.org/10.1017/S001447971700028X

[9] Vayssieres, J.-F., Korie, S., Coulibaly, O., Temple, L., & Boueyi, S. P. (2008). The mango tree in central and northern Benin: Cultivar inventory, yield assessment, and loss due to fruit flies. Fruits, 63(6), 335–348. https://doi.org/10.1051/fruits:2008035 Vila et al. (2001). [Forest growth and density studies]. (n.d.).

[10] Djaha, A. J.-B., N’Da, H. A., Koffi, K. E., & Adopo, A. N. (2014). Diversité morphologique des accessions d’anacardier (Anacardium occidentale L.) introduits en Côte d’Ivoire [Morphological diversity of cashew accessions introduced in Ivory Coast].

[11] Diatta, B. S. (2019). Caractérisation écologique des parcs agroforestiers àAnacardium occidentale L. dans le Département de Goudomp (Région de Sédhiou / Sénégal).

[12] Ndour, K., & Faye, E. (2022a). Caractérisation structurale, morphométrique et phénotypique de trois morphotypes d’anacardiers (Anacardium occidentale L.) au Sénégal [Structural and morphometric characterization of cashew morphotypes in Senegal].

[13] Hullat, L. (2024). Méthode des moindres carrés ordinaires : Statistiques, régression [Ordinary Least Squares Method]. Studysmarter FR. https://www.studysmarter.fr/resumes/economie-etgestion/economie-manageriale/methode-des-moindres-carres-ordinaires/ Kantousan. (2019). [Master’s Thesis/Memory].

[14] Ndiaye, S., Charahabil, M. M., & Diatta, M. (2017). Caractérisation des Plantations à base d’anacardier (Anacardium occidentale L.) dans le Balantacounda : Cas des communes de Kaour, Goudomp et Djibanar (Casamance/Sénégal). European Scientific Journal, ESJ, 13(12), 242. https://doi.org/10.19044/esj.2017.v13n12p242

[15] Banla, T., Banito, A., Houehanou, T. D., Kpémoua, K. E., Songai, S. M., Tchalla, P., Kokou, K., Koba, K., Sanda, K., & Kadai, E. A. (2023). Caractérisation agronomique des anacardiers (Anacardium occidentale L.) élites sélectionnées dans les zones de production au Togo. 42.

[16] Elhadji Thierno, N. (2024). Effets de différents substrats de sol sur la levée de germination et la croissance de variétés d’anacardiers (Anacardium occidentale L.) au niveau de la station de l’ISRA de Sangalkam/Sénégal. Revue Ecosystèmes et Paysages, 4(2), 1‑15. https://doi.org/10.59384/recopays.tg4206

[17]  Yéo, L. (2015). Figure 1 : Carte de la région du Poro (Côte d’Ivoire) [Map of the Poro region]. ResearchGate. https://www.researchgate.net/figure/Carte-De-La-Region-Du-Poro-Cote-Divoire_Fig1_333094909