Desert Locust Decision Support for Improved Agriculture Production

Abstract

The desert locust (Schistocerca gregaria) is widely considered to be the most dangerous migratory pest species due to its rapid reproductive capacity, long-distance migration potential, and devastating impact on agriculture and ecosystems. In addition, desert locust populations have increased rapidly, and swarms have invaded eleven countries in West Africa, severely disrupting agricultural production in vulnerable areas already facing security challenges. Timely and accurate information on the desert locust through remote sensing is critical for effectively managing and improving agricultural production, especially in West Africa, where such information is scarce. The objective of this work is to enhance the monitoring and prevention efforts against desert locust outbreaks by integrating remote sensing and decision-support tools. The tool identifies locust development and gregarization zones in order to assess the risk of outbreaks and support decision-making processes. It combines a model on the presence or absence of transient phases of the species, biotope ecological conditions, and gregarization thresholds for both juveniles and flying adults. In this paper, the Google Earth Engine platform is used to monitor eco-meteorological conditions in key desert locust survival and breeding areas using high-resolution geospatial data. The improvement initiative covers certain aspects of the user interface, real-time data updates, and a fully operational set of eco-climatic indices impacting locust multiplication. These advancements contribute to a more robust decision-support system for locust early warning and control in West Africa.