Trailing smart cocoa farming technologies

12 December 2019

Precision agriculture technologies for the cocoa industry in West Africa and Indonesia is often deemed far-fetched. Smallholder farmers in these regions, however, are susceptible to negative changes in the environment and often lack timely, site specific and actionable advice.

FarmGrow continues to trial new technologies that have the potential to bring a greater control over the production process and farm management for cocoa farmers and their coaches. Therefore, we collaborate with Waterwatch Cooperative, an ag-tech organization focusing on optimizing agricultural production with efficient use of resources by applying tech-based services.

For FarmGrow, Waterwatch has trailed a service based on drone technology to support the field activities of the coaches. FarmGrow coaches visit farmers several times a year and give detailed advice to the farmer on how to adjust cocoa farm and cultivation practices in order to increase the productivity. The drone-based service provides detailed insights on the farm layout and agronomic conditions that support the coach and farmer in making good decisions.

“The service provides detailed insights on the farm layout and agronomic conditions that support the FarmGrow coach and farmer in making good decisions.”

For example, the high-resolution drone images provide the coach and the farmer with a detailed overview of the farm from above. These images enable the coach to make a very targeted farm inspection. On these images you can see the layout, pattern and quality of cocoa trees, shade tree cover, dead trees, open spots and other vegetation as palm oil and plantain. This information gives a quick overview on the setup of the farm and which areas need attention. Both coaches and farmers indicate that this is helpful to get a good overview of the of the problem areas in the farm and to setup the inspection more efficiently, which saves time in the process.

To even better understand the farm, two so called masks have been developed based on the drone imagery. These masks can highlight and distinguish certain properties and land use on the farm in high detail. The first masks distinguishes the locations of cocoa trees from other crops and vegetation types and points out where there are still open spots (see brown color in the image below). The second mask shows the location of shade trees and determines the height of these shade trees (see blue colour in the images below). This information can be used to quickly assess the farm setup and discover areas that have a potential for denser cocoa planting, and more optimal use of shade cover.

High resolution areal images from a cocoa farm in Sefwi Bekwai, Ghana. On the first and second images, shade trees are displayed on a colour range from dark blue (5-9m) to light green (30-35m) depending on tree height. The cocoa classification (second image) is displayed in a brown colour, showing the location of cocoa trees.

More stories