Project

Crop ecological aspects and modelling of cocoa production

Using a mix of experimental and 3D modelling approaches the project aims at understanding pruning effect on cocoa tree functioning and production.

Background

Cocoa is an important commodity crop and source of income for millions of small holder farmers. Over 70% of word cocoa production is concentrated in West Africa, with  Côte d’Ivoire being by far the lead producing country. Production increased over the past decades thanks to a strong expansion of cocoa cultivated area, often at the expenses of forests, while yield per hectare remained low. Low yields are due to several cooccurring factors, such as low soil fertility, aging plantation, and inadequate maintenance practice.  Pruning is considered an essential yield-enhancing practice but its effects on cocoa growth and yield are poorly understood and pruning recommendation lack scientific underpinning.

Project description

The project aims at better understanding the effect of pruning on cocoa architecture, growth and production.  

To achieve this we used a combination of experimental and modelling approaches. We set up 3 experiments. In a first experiment we looked at cocoa branching pattern in young plants and how branching is influenced by different pruning interventions (Experiment 1). In a second experiment we  tested various formation pruning strategies ( the pruning intervention carried out at young stage to shape the tree crown) in two shade level conditions, assessing how pruning influenced tree growth (Experiment 2). In the last experiment we studied the effect of a recommended pruning strategy on light capture, flushing and cocoa pod production in 12 year old cocoa trees (Experiment 3).

We also assessed how this effect were mediated by tree size and competition with neighboring cocoa trees. We then used many of the information gathered in our experiments to develop and calibrate a functional structural plant (FSP) model of juvenile cocoa trees. The model simulates cocoa growth and development and it includes architectural responses to pruning intervention. The model serves as a tool to aim the development of more efficient pruning practice, tailored to different shading and density regimes. In addition it can be used for screening promising trial options and for ideotyping plant structure for breeding purposes. The 3D visual output makes the model a communication tool for academic and non-academic audience. 

Results of experiment 3

After one year, pruned trees recovered initial losses in whole-canopy light interception but maintained a more uniform distribution of light in the canopy. Pruning directly increased flushing activity. Pruning enhanced the positive effect of tree size and strongly reduced the negative effect of competition on pod number. The predicted net effect of pruning on the number of harvested pods varied greatly with tree size and competition, ranging from -58% for small trees under low competition, to +150% for large trees under high competition. This large variability stresses the importance of individual-level analysis to quantify pruning effects and calls for more attention to individual tree characteristics in training and practice of cocoa pruning. Additionally, the mitigation of competition effects by pruning suggests that adequate pruning may help to realize cocoa stands with a higher planting density. 

Results experiment 3
Results experiment 3

FSP cocoa model

The model is under development. Here below a glimpse of its visual 3D output.

FSP cocoa model
FSP cocoa model

Publications