Leveraging the EO4EU platform, SISTEMA has developed a robust, end-to-end pipeline capable of collecting, processing, and feeding satellite and ground data into an advanced machine learning model to deliver actionable insights. For this use case, SISTEMA created a hybrid model that combines a Recurrent Neural Network (GRU) with Maxent, a powerful statistical technique, to predict the early development stages of Desert Locusts, namely Hoppers and Bands. The model was trained and validated using FAO Locust Watch dataset, historical locust data from 2000 and 2021.

By automating the entire data lifecycle, from satellite data collection to model-driven forecasting, this pipeline helps users to make smarter, faster, and more cost-effective decisions. 

Techical Details 

Overview

The proposed use case represents an advanced information service for assessing and predicting the impact of Desert Locust plagues, designed to support NGOs, international institutions, and decision-makers with timely, data-driven intelligence. Building on a successful pilot in collaboration with an NGO operating in Yemen, SISTEMA is enhancing its existing damage assessment tools by integrating a predictive component, empowering stakeholders to shift from reactive response to proactive prevention.
Desert Locust outbreaks are closely tied to dynamic environmental factors such as climate conditions and vegetation health. Traditionally, Earth Observation (EO) data and climate forecasting have been analyzed in isolation. SISTEMA’s innovative approach combines both datasets using cutting-edge AI and machine learning techniques to deliver a more accurate and comprehensive prediction model.

This next-generation service introduces:

• Conceptual advancements through the integration of new climate models for improved forecasting.
• Technical upgrades powered by High-Performance Computing (HPC) infrastructure, ensuring scalability, speed, and performance.

The result is a robust, integrated solution that enhances locust monitoring and early warning capabilities, enabling faster, smarter, and more effective interventions on the ground.

Challenge

The Desert Locust (Schistocerca gregaria), found in various parts of Africa, Asia, and the Middle East, is considered the most destructive migratory pest in the world because it is highly mobile and can form swarms containing millions of locusts, leading to devastating impacts on crops, pasture, and fodder. A small swarm (1 km2) can be made up of 80 million locusts and can consume the same amount of food in one day as 35,000 people, while a large swarm can eat up to 1.8 million metric tons of green vegetation, equivalent to enough food to feed 81 million people. Without broad scale control, conservative estimates for locust-related losses including staple crops, livestock production and asset damages are estimated at US$8.5 billion for countries in the wider East Africa region, Djibouti and Yemen.
In 2021, the Director-General of the Food and Agriculture Organization of the United Nations (FAO), QU Dongyu, has welcomed an €8 million contribution from the European Union (EU) that will enable countries in the Horn of Africa to maintain a state of readiness in the face of a possible resurgence of Desert Locusts.
To improve this kind of services, the contribution of climate forecast is fundamental to increase the reliability of climate impact prediction and to provide reliable risk information to local operators and to government institutions.

Solution

The Desert Locust monitoring and prediction service that addresses both early warning and real-time crisis management, delivering critical intelligence to support timely, effective responses. This dual-response approach includes:

• Early warning, to prevent the up surging of the locusts: this answer is mostly preventive by identifying areas that shows good environmental conditions for the locust breeding.
• Situational, to follow the crisis already going on, identifying the swarms and their impact to provide an efficient humanitarian answer.

The EO4EU platform helps monitor and control desert locusts, which are insects that can cause serious damage to crops and threaten food security in many regions.
The following steps has to be follow in order to work with desert locust monitoring service on EO4EU platform:

• Collecting Data

EO4EU gathers different types of information, such as:

• Satellite datasets
• Bioclimate model datasets

 

• Combining the Information

The platform fuses all these datasets into one system. This helps create a clearer and more complete picture of the locust situation.

 

• Analyzing the Situation

With the combined data, EO4EU can map and show where desert locusts are now

Figure 1: Use case 6 workflow on EO4EU platform

The input variables are:

• Soil water content layer 1 (ERA5-Land), up sampled with nearest neighbour to 5km
• NDVI (MODIS), original at 5km
• Total precipitation (ERA5-Land), up sampled with nearest neighbour to 5km
• Temperature (ERA5-Land), up sampled with nearest neighbour to 5km

The model takes as input a time series, for a single point, of 50 days previous the current date. The 50 days are divided into 10 time-steps each corresponding to a 5 days average. MODIS provide NDVI data every 8 days, so data is interpolated temporally with using bilinear algorithm.

Figure 2. Extraction of time series of climate variables on points where Desert Locusts have been reported by FAO.