NJIT Biologist Among $10M XPRIZE Competition Winners for Rainforest Biodiversity Sampling Tech

Newark, N.J., Nov. 18, 2024 (GLOBE NEWSWIRE) -- NJIT biologist Eric Fortune and a team of scientists called “Limelight Rainforest” have won the five-year XPRIZE Rainforest Competition, securing half of the competition's $10 million prize purse. 

The team's dramatic victory was announced Nov. 15 at the G20 Social Summit in Rio de Janeiro, the culmination of a global competition that began in 2019 when the nonprofit XPRIZE Foundation challenged innovators around the world to “develop technology to capture the true biological diversity of rainforests…and show the value of protecting the natural resources within them.”

Ultimately, Limelight Rainforest and its biodiversity sampling technology, “Limelight”, outshined the competition — topping an initial field of 300 teams from 70 countries, including six finalists that competed in Amazonas in Manaus, Brazil this past July.

“It's amazing. Being part of this crazy adventure over three years has been an enormous learning experience with so many twists and turns. It’s something I'll never forget," said Fortune, a team leader who developed sensor and control systems used in the Limelight data collection platform.

"The real reward is that this work can have a lasting impact on these vital ecosystems and the communities that depend on them. It’s what drew us to this competition in the first place.”

Originally started by Colorado Mesa University biology professor Thomas Walla, Limelight Rainforest forged its drone-based sampling technology under the unique rules of the competition's various stages held in rainforests around the world — teams have been challenged to deploy technologies to remotely survey as much biodiversity as possible across 100 hectares of forest accurately, and in under 24 hours.

VIDEO: XPRIZE Rainforest Finals Competition

The Limelight team of more than 50 engineers, biologists and indigenous scientists developed their system to collect a flurry of data on species that included ultrasound produced by bats and traces of DNA left by primates moving through the forest.

The latest Limelight uses telemetry, satellite communications and advanced AI to interpret the hundreds of thousands of images, recordings and samples collected by the platform’s advanced microphones, cameras and capture systems. The 24-hour sampling period was followed by a 48-hour sprint to produce a final report of species, their movements and deep insights into the forest's biodiversity.

Uncovering Life in the Amazon

On the heels of the semifinals hosted in Singapore, the XPRIZE Rainforest Competition finals were fittingly held in the Amazon — the most biodiverse place on Earth, estimated to be home to over 10% of the known species in the world.

“We were taken by boat to this remote location on the shores of the Rio Negro where they had a hut for us to spend the next 24 hours,” Fortune said. “Our team has a lot of experience in the Amazon, so we were confident, but we didn’t expect it to go so smoothly.

“We sent out 10 Limelight rafts with our drones once they put us on the clock, and from there everything just worked.”

Upon deployment, the team's devices — each equipped with five light trap cameras — lured and imaged an astonishing 250,000 insects that were classified in mere hours.

The Limelight’s new water sampling tool, deployed using custom robotic systems, filtered 45 liters of water from remote streams in narrow canopy gaps to catalog the vast biodiversity of the forest’s aquatic ecosystem. It yielded over 27 million environmental DNA (eDNA) sequencing reads on location.

The team also pioneered a new “Nature Node” acoustic system, capable of identifying bird and other tree-dwelling species by their vocalizations with unprecedented precision.

“For 40 years, people have been trying to track animals based solely on their vocalizations. It was a dream of mine as a grad student and our team made it happen,” Fortune noted.

VIDEO: Finals Testing Insect Timelapse

AI Trained by Indigenous Experts

One of Limelight’s biggest standout features in the finals was the accuracy of their AI in identifying the vast array of forest life.

For that, the team turned to the Quechua and Waorani Indigenous groups native to the rainforests of Ecuador, who reviewed and validated thousands upon thousands of sounds and images of rainforest species. Part of this effort was funded by a Kickstarter campaign to train their AI’s species identification capabilities.

“Many of our team members have been conducting research in Ecuador and building relationships with the Quechua and Waorani groups for over 20 years,” said Fortune. "Our Indigenous team members are the true masters of this knowledge. They were vital in helping confirm the identifications of these species … in many ways the team was completely reliant on their expertise.

“We were certain our AI was trained well because we had the world's top experts validating the data that we fed into the AI.”

Upon its XPRIZE success, the team is already planning to scale up Limelight production. Fortune is helping lead development of the first generation of Limelight devices for real-world use, partnering with NGOs, Indigenous communities and other organizations invested in rainforest conservation.

The experience is one he will not soon forget, but Fortune says the team’s work toward rainforest conservation has only just begun.

 “We're already identifying projects in South America and Southeast Asia where we could have an impact, redesigning our systems for these real-world needs,” said Fortune. “Once these devices start to be deployed, we will make enormous discoveries and uncover so many hidden organisms unique to these rainforests. It could change how we value and protect them.”

Attachments


Deric Raymond
New Jersey Institute of Technology
9736427042
draymond@njit.edu
Data & News supplied by www.cloudquote.io
Stock quotes supplied by Barchart
Quotes delayed at least 20 minutes.
By accessing this page, you agree to the following
Privacy Policy and Terms and Conditions.