Traditionally, taking inventory of the species in a rainforest requires sending in a team of experts with field guides and binoculars for a multi-day expedition. But the devastating pace of the destruction of the world’s rainforests and increasing urgency to better monitor and protect what remains demand faster, easier, and more efficient approaches.
Several years ago, a Yale-based team devised an alternate approach: they use lightweight, unmanned aerial vehicles (UAVs) to collect this critical biodiversity data in remote areas.
Now they’ve collected something else: a coveted international honor.
XPRIZE Rainforest, a $10 million global competition to find the most innovative technology for exploring Earth’s biodiversity, has awarded one of its top prizes to Map of Life Rapid Assessments (MOLRA), an international research group led by Walter Jetz, professor of ecology and evolutionary biology in Yale’s Faculty of Arts and Sciences and director of the Yale Center for Biodiversity and Global Change.
The MOLRA team placed second in the five-year competition, earning a $2 million. XPRIZE Rainforest officials made the announcement Nov. 15 at a ceremony associated with the G20 Social Summit in Rio de Janeiro.
“We are immensely excited about this recognition,” said Jetz, who led in the creation of the groundbreaking Map of Life platform more than a decade ago. “In the face of rapid biodiversity loss, more accessible and effective tools to measure and plan for biodiversity are urgently needed. We are beyond grateful as the award will allow us to grow our solution that can quickly and cost-effectively deliver actionable biodiversity insights for locations anywhere.”
Map of Life is a global database that tracks the distribution of known species. It is now used by world leaders to monitor, research, and create policies that protect species worldwide.
MOLRA uses the Map of Life database as an engine to combine biodiversity research, innovative survey technologies, and cutting-edge informatics tools to deliver comprehensive local biodiversity information and support conservation action. The company’s fleet of semi-autonomous UAVs collect audio, visual, and environmental DNA samples with minimal human intervention.
The MOLRA XPRIZE team included members from the Yale Center for Biodiversity and Global Change, the Field Museum of Natural History, the Rutgers Environmental DNA Lab, the Federal University of Amazonas, Trinity University, the University of East Anglia, and Sony Group Corporation.
The five-year XPRIZE Rainforest competition — which seeks to enhance mankind’s knowledge of the rainforest ecosystem by highlighting innovative technology that expedites the monitoring of tropical biodiversity — began in 2019 with 300 teams from around the world.
In July, the competition’s six finalist teams were asked to survey 100 hectares of tropical rainforest near Manaus, Brazil, in 24 hours. Their task was to produce meaningful, real-time insights from their data within 48 hours. In addition, each team had to demonstrate the scalability of their technology.
The MOLRA team recorded 225 species from 5,500 individual, geolocated identifications, from anteaters to palm trees to frogs — thanks to a fleet of drones that were mostly pre-programmed to fly missions through the canopy. They collected more than 4,000 photos, 26 hours of audio recordings, and 24 eDNA samples (traces of DNA left by organisms in soil, water, and in the air).
The team was able to identify species from raw samples because of its advanced new modeling technology, state-of-the-art artificial intelligence (AI) algorithms, innovative eDNA processing techniques, and collaboration with biodiversity experts in Brazil and all over the world. Four of the recorded species are globally threatened with extinction, including the giant anteater, the yellow-footed tortoise, the ringed woodpecker, and the white-crested guan.
In the semifinal round, in 2023, MOL Rapid Assessments identified more than 150 species in the central rainforest of Singapore. This came after accumulating 2,199 visual samples, 292 acoustic samples, and 1,419 species identifications of plants, mammals, birds, reptiles, amphibians, and insects.
Several factors contributed to the MOLRA team’s success in the XPRIZE Rainforest competition, Jetz said. First, it leveraged the Map of Life to predict what species might be found at any site around the world — helping to guide sampling design and the use of AI. Second, no specialists were needed to operate the system on the ground thanks to MOLRA’s largely autonomous nature. Finally, MOLRA’s combination of AI and human biodiversity experts optimizes the breadth and accuracy of species identifications.
MOLRA will use the prize money to hire staff and expand its work around the world, Jetz said. With support from Yale Ventures (a university initiative that supports innovation and entrepreneurship campuswide), the team is working with partners across domains and sectors to deliver effective and accessible biodiversity measurement solutions.
“Our goal is to deliver a scalable, easy to use, and low-cost biodiversity assessment solution that empowers local stakeholders to protect the natural places they depend on — at the speed we need to meet the world’s ambitious biodiversity goals,” said Nigel Pitman, a botanist at the Field Museum in Chicago who coordinated the MOLRA plant inventory.
“The technological advances and automation of biodiversity surveys is a major step towards being able to catalog the distribution of the Earth’s biodiversity efficiently and effectively,” said Izeni Farias, a member of the MOLRA team from the Federal University of Amazonas in Brazil. “The MOLRA tools are accessible and simple to use and can be used to deliver immediate conservation insights to local communities and strategic partners, resulting in meaningful conservation actions.”
The MOLRA team drew on the expertise of more than a dozen taxonomic experts from around the world and received crucial support from Esri (Environmental Systems Research Institute), Sony, Google, and the E.O. Wilson Biodiversity Foundation.
# # #