Armyworm research using remote-sensing methods

The Fall Armyworm, the caterpillar stage of a moth, is a major pest in North and South America which crossed the Atlantic to West Africa in 2016 and rapidly spread across the continent. This project, led by Gorta-Self Help Africa, developed an Artificial Intelligence (AI) algorithm to detect the damage caused by Fall Armyworm in maize fields in Balaka District, Malawi. It addressed the following SDGs: SDG 2 - Zero hunger, SDG 12 - Responsible consumption and production, SDG 13 - Climate Action, and SDG 15 - Life on land.

The Fall Armyworm feeds on over 80 plant species but prefers maize, which is the staple food crop for much of East and Southern Africa, causing up to 30% crop losses. The project succeeded in developing a machine learning algorithm that could detect the damage caused by the Fall Armyworm (FAW) to smallholder maize fields using satellite images. The software can detect the presence or absence of FAW damage with up to 87% accuracy and the level of FAW damage with 63-75% accuracy, depending on the data available. The success of this “proof of concept” research provides the foundation for the development of remote sensing software that can identify FAW damage hot spots at the landscape level, enabling governments to target scarce control resources on these hotspots. The success also indicates that this approach can be used to detect other crop pests and diseases that change the color of the crop canopy.

For example, the lessons learned from this project have been applied in the extension of the Fall Armyworm (Spodoptera frugiperda) research project funded by the US Foundation for Food and Agriculture Research (FFAR) and in recent proposals to adapt the software to detect damage caused by Desert Locusts (Schistocerca gregaria) and Banana Bunchy Top Virus.

Final report

Lessons Learned

Disclaimer: As many of the links are external, some links may not work.