“The Olympics for biodiversity” just wrapped up in Brazil. Abhishyant Kidangoor provides an overview. In July, the Amazon rainforest became the testing ground for 6 teams competing for the $10 million Rainforest XPRIZE, aimed at developing innovative solutions to automate biodiversity monitoring. The competition, organized by XPRIZE, brought together biologists, engineers, and AI experts to put their technologies to the test over a 72-hour period. Each team, tasked with collecting and analyzing biodiversity data, produced a comprehensive report on the species they identified using a range of technologies with a focus on affordability, scalability, and collaboration with local communities. The winner will be announced soon. Kidangoor profiles the finalists: 🌳 Limelight Rainforest deployed drones to drop rafts into the canopy. Equipped with cameras, microphones, and light traps, these rafts attracted insects for DNA sampling. Air samples were also taken for eDNA analysis, while a robotic drone collected leaf and water samples. Machine-learning models mapped tree canopies, estimated tree diversity, and measured carbon storage. 🌳 Welcome to the Jungle used biodegradable sensor packages to collect data. Drones with multispectral and lidar cameras mapped forest cover and gathered eDNA samples. Machine-learning models processed this data, identifying species from both images and sounds, with a focus on forest health. 🌳 Brazilian Team divided the forest into 24 clusters, using drones and the Pl@ntNet AI platform to identify tree species. Drones with thermal cameras detected wildlife at night, while canopy sensors recorded insect activity. Water and plankton samples were also collected. 🌳 ETH BiodivX deployed canopy rafts equipped with light traps and sticky tapes, using drones to collect eDNA samples. Aluminum-based devices analyzed the samples, detecting traces of malaria from monkey feces. Machine-learning models identified bird species, insects, and mapped the forest using RGB cameras. 🌳 Providence Plus used their autonomous Deep-Rainforest Operation Platform (DROP) to collect and process sound, images, and eDNA data from canopies and water bodies. Motion sensors enabled real-time photo capture, with AI identification models. Drones collected soil and water samples, which were analyzed for eDNA. 🌳 Map of Life Rapid Assessments focused on a software platform, utilizing autonomous drones to collect images, audio, and eDNA data. The tools, connected with the Map of Life database, enabled autonomous surveys, species identification, and 3D canopy mapping. Real-time analysis and species assessment were conducted through remote collaboration, merging local data with global biodiversity datasets. 📰 At the ‘Biodiversity Olympics,’ scientists work to democratize rainforest tech https://lnkd.in/gnd3Nt2f
-
+4