One Forest Vision: Better Understanding Tropical Forests to Better Protect Them
The One Forest Vision initiative harnesses research, remote sensing, artificial intelligence, and citizen science to enhance knowledge and monitoring of tropical forests. In the Congo Basin, it relies notably on Pl@ntNet to facilitate plant identification, document poorly known floras, and build local capacity.
Tropical forests harbor exceptional biodiversity and store vast amounts of carbon. They play a vital role in regulating the climate and the water cycle while providing numerous resources to local communities. However, they are threatened by deforestation, resource exploitation, wildfires, agricultural expansion, infrastructure development, and climate change.
Protecting these ecosystems requires accurate information on species, forest structure, and their carbon storage capacity. Yet, in many tropical regions, ground-based data remain scarce or scattered across different institutions.
Launched in 2023 during the One Forest Summit in Libreville, One Forest Vision (OFVi) supports scientific cooperation with countries located in major tropical forest basins. Its initial deployment focuses on the Congo Basin, particularly in Gabon, the Republic of the Congo, and the Democratic Republic of the Congo.
Connecting the Field with Space-Based Observation
OFVi combines multiple methods for studying biodiversity and carbon:
- Botanical and forest inventories;
- Long-term monitored permanent plots;
- Tree, biomass, and soil measurements;
- Camera traps and acoustic recorders;
- Drones and LiDAR technology;
- Satellite imagery;
- Ecological models and artificial intelligence.
Long-term research sites, known as “super-sites,” act as open-air laboratories. Scientists simultaneously study vegetation, fauna, ecosystem functioning, and the carbon cycle there. Ground-based observations help ground-truth, calibrate, and improve the information generated by satellites and drones.
This synergy is essential: satellites provide regular monitoring over vast areas, while field inventories accurately identify species and measure the specific characteristics of each forest.
Pl@ntNet to Facilitate Tropical Plant Identification
Plant identification is one of the main challenges in tropical surveys. A single forest can be home to several hundred tree species, some of which are rare, visually similar, or poorly documented. Specialized botanists are indispensable, but far too few to cover such immense territories on their own.
One Forest Vision relies on Pl@ntNet to help unlock this “taxonomic bottleneck.” The platform is continuously enriched with images from herbaria, scientific inventories, and new field campaigns. Once accurately identified and validated, these data are used to train and improve automated recognition models.
For Pl@ntNet, the initiative helps to:
- Better represent the floras of the Congo Basin;
- Integrate expert-validated photographs;
- Enhance the identification of rarely observed tropical species;
- Generate new geolocated observation records;
- Expand the platform’s applications for scientific research and conservation.
Ground-level photographs can also complement canopy images captured by drones. Combining both offers new possibilities for identifying trees across different spatial scales and mapping forest composition.
Training and Engaging Local Stakeholders
Capacity building is a cornerstone of the initiative. Training sessions on Pl@ntNet are organized in collaboration with universities, research institutes, herbaria, botanical gardens, and protected area managers.
In Gabon, field missions conducted in 2025 and 2026 trained students, educators, researchers, and field guides. Observations were cross-referenced with reference collections preserved in local herbaria.
These activities demonstrate that Pl@ntNet is more than just an identification tool: it is also an effective platform for training, knowledge sharing, and building local botanical resources.
Observations That Serve Conservation
A Pl@ntNet observation links photographs with a date, geographic coordinates, and a taxonomic identification. Aggregated at a regional scale, these data can help map species distributions, document rare or endangered plants, detect invasive species, and track flowering or fruiting cycles (phenology).
They can also assist in characterizing habitats, prioritizing inventories in understudied areas, and analyzing the impacts of forest degradation and climate change. In this way, Pl@ntNet data seamlessly complement ecological field measurements, satellite imagery, and drone-collected data.
More info at https://www.oneforestvision.org/eng
