Drivers of cocoa yield and growth in young monoculture and agroforestry systems

Abstract

CONTEXT

The expansion of cocoa farming has been linked to deforestation and biodiversity loss in West Africa. Agroforestry systems could potentially increase the sustainability of cocoa production. However, despite the long history of cocoa cultivation in agroforestry systems, the exact mechanism or the combination of factors that drive cocoa growth and yield in these agroecosystems, especially at the cocoa establishment stage, is unclear.

OBJECTIVE

The present study aims to analyse how resource availability, stand characteristics and cocoa tree morphology interact to determine the performance of cocoa during the establishment phase among different cocoa cultivation systems.

METHODS

We studied seven different agroforestry systems and one monoculture system in 53 plots located in a 12-ha cocoa agroforestry trial recently established in Côte d'Ivoire. We characterized each system with 16 variables that described five system components: 1) cocoa yield (number of pods, pods weight, number of productive trees) and cocoa growth (basal diameter), 2) cocoa tree morphology (crown depth, diameter, area, and volume), 3) stand characteristics (number and size of shade trees), 4) light (leaf area index) and 5) water (soil volumetric water content). We used a Structural Equation Modelling (SEM) approach to understand and quantify interactions between the five components.

RESULTS AND CONCLUSIONS

The models were able to largely explain cocoa early yield and juvenile growth variability (respective determination coefficients: 0.45 and 0.92). We observed improved cocoa establishment in certain designs of agroforestry systems compared with monoculture. However, SEM revealed that stand characteristics (i.e. increase in the number and height of shade trees) had a negative effect on cocoa yield and growth via changes in cocoa tree morphology, having path coefficients of −0.48 and − 0.72, respectively. Conversely, we found a positive relationship between the cocoa tree morphology and cocoa early yield, for which the path coefficient was 0.74, the strongest relation. Cocoa tree morphology was more important than stand characteristics, leaf area index, and soil volumetric water content in controlling cocoa yield and growth. Taken together, our results suggest a competition for space between shade trees and cocoa plants. Finding the optimal design of agroforestry systems can help enhance cocoa establishment in the first years, improving the long-term performance of the system.

SIGNIFICANCE

This finding suggests that the competition effect depends on spatial planting pattern and species composition, offering an opportunity to meet the goals of a sustainable intensification of cocoa plantations through the design and management of optimal agroforestry systems.