Subscribe
As locusts become one of the most dangerous migratory pests for millions of smallholder farmers and a threat to food security, scientists have developed a tool to forecast desert locust populations and their short- and long-term swarm movements.
The period of 2019–2021 saw one of the largest desert locusts upsurges in two decades. Locust invasion extended from Kenya to India, causing large-scale devastation to crops and general vegetation.
Scientists from the University of Cambridge, United Nation’s Food and Agriculture Organization (FAO) and United Kingdom’s Met Office, have developed an integrated modelling framework to predict the behaviour of migratory locust populations.
Using a combination of weather and environmental data together with an atmospheric transport model, the tool can forecast swarm movements and the spread of desert locusts.
Locust invasions are dramatic and sudden and during upsurges, they form massive swarms that can travel long distances and destroy every vegetation that comes in their way.
With this tool, the scientists aimed to provide a practical resource and a starting point to government agencies, agronomists and farmers to prepare for and respond to the next upsurge.
The framework was developed by integrating:
• selection of locusts breeding sites
• maturation through egg
• hopper and adult stages
• feeding and swarm dispersal in search of areas suitable for feeding and breeding
• incorporation of weather-driven models for wind trajectories to predict daily pathways of swarm migration
• use of remote-sensed data to predict the duration of swarm feeding at a single location given the state and availability of vegetation for feeding
• assessment of flight direction, likely origin, and landing sites for swarms observed in flight
The new tool, with an accuracy of 83 per cent, can be crucial in predicting breeding areas that may be missed during ground monitoring.
The study area for developing the model extended across five sub-Saharan countries (Kenya, Ethiopia, Somalia, Eritrea and Djibouti) that were affected by the 2019 locusts upsurge. It was tested against the pattern of recorded swarm movement in these areas.
“The model also accounts for the feeding behaviour of desert locusts, where the duration swarms spend at a landing site is determined by the availability of food at the site. Our aim was to provide a cohesive and comprehensive framework based on the available data and current understanding of desert locust biology,” noted the authors of the paper published in December 2024 in the journal PLOS Computational Biology.
Scientists said that previous modelling studies on desert locusts did not extend to accounting for the migratory behaviour of desert locust swarms that arise from h opper bands. The new tool has addressed that and integrated not only feeding and movement of swarms, but the whole life cycle of desert locusts with environmental conditions.
For developing the framework, the scientists extracted land cover data for land cover classification from the Copernicus global map of land cover at 100 m resolution. Land cover maps represent spatial information on different types of the earth’s surface eg. urban areas, sparse vegetation, cultivated vegetation.
Next, they prepared a breeding suitability map by identifying breeding areas and date, followed by checking if the environmental conditions at the specified location are suitable for egg laying.
After calculating an egg incubation period leading to hopper emergence, the scientists tested if conditions in terms of available vegetation were suitable for locust feeding.
The final step was dispersal and feeding of swarms, which involved analysis of wind-assisted dispersal trajectories and testing how long vegetation available on a ground can sustain a swarm at successive landing sites.
India suffered one of the worst locust attacks in the history in 2020-21, when desert locusts migrated to the country after a gap of 26 years and raided several districts in Rajasthan, Gujarat, Haryana, Punjab, Uttarakhand, Uttar Pradesh, Bihar, Chhattisgarh, Madhya Pradesh and Maharashtra.