Deployment and Evaluation of the Autonomous Monitoring of Insects (AMI) System for Ecosystem Health Assessment in Kenya

Class Insecta make up about 40 % of the terrestrial animal biomass. Due to biomass size and incredible adaptability to various stresses in the environment such as changes in temperature and predators, insects can be used as ecosystems health indicator species when closely monitored. This paper details the deployment of an autonomous insect monitoring system, and the development and implementation of an image-processing based insect count algorithm. The Autonomous Monitoring of Insect (AMI) system was developed by the UK Centre for Ecology and Hydrology and deployed at the Dedan Kimathi University Wildlife Conservancy in Kenya. The insect-count algorithm was able to count insects, averaging 10 per night over two months, with peaks between 8.00 pm and midnight. Initial challenges included images with overlapping insects, which affected accuracy, highlighting the need for more refined count algorithms. Future work will expand to species classification and behavioural analysis, particularly moths. By providing continuous and autonomous data, the AMI system offers a scalable tool to assess biodiversity trends which inform ecosystem management in Kenya.