Biodiversity monitoring faces significant challenges across Europe, from the labour-intensive nature of traditional field surveys to the shortage of taxonomic expertise needed to identify species. The EU-funded project MAMBO (Modern Approaches to the Monitoring of Biodiversity) addresses these challenges through innovative monitoring technologies that combine artificial intelligence, remote sensing, and automated field equipment, transforming how biodiversity is assessed across Europe.

MAMBO has developed a set of technologies designed to monitor different components of biodiversity. The AMI-trap is an autonomous light-trap system that captures high-resolution images of nocturnal insects using UV and LED light sources with motion-detection software. Capable of processing over one million recordings per season per device, the system has been deployed in more than 30 countries. For diurnal pollinators, insect camera traps record images of flowering plants and visiting insects, classifying up to 80 taxa groups at genus or species level with over 100,000 recordings per season per camera trap.

Sound recognition models identify species from audio recordings across four European animal groups: breeding birds, amphibians, bats, and grasshoppers. Compatible with equipment from smartphones to professional recorders, the models analyse recordings in three-second time windows, enabling well-standardised data collection across multiple species groups using a single method. For citizen science, the Multi Source Model (MSM) provides image recognition for over 41,000 taxa of European animals, plants, and fungi. Trained on 35 million validated images and deployed across over ten biodiversity portals, the model performs over 76 million identifications annually, with records uploaded to GBIF after validation.

The Plant Quadrat Image analysis tool, available on the Pl@ntNet platform, transforms a photograph of a 50 × 50 cm plot into a structured description of the entire plant community present, delivering plot-level results in seconds. At European scale, high-resolution species and habitat maps at 50 × 50 m spatial resolution describe all plant species and EUNIS habitats, made interactively explorable through the GeoPl@ntNet platform. The Airborne Laser Scanning tool uses Laserfarm, an open-source Python workflow, to convert LiDAR point clouds into standardised map layers capturing vegetation structure, applied to seven demonstration sites and producing full national coverage for the Netherlands.

For more information on all technologies developed within MAMBO, please visit the project’s website.

—

This project receives funding from the EU Horizon Europe Research and Innovation Action programme under Grant agreement No. 101060639.

Views and opinions expressed are those of the author(s) only and do not necessarily reflect those of the European Union or the European Commission. Neither the EU nor the EC can be held responsible for them.