Team wins $10M XPRIZE Rainforest competition for novel solution

Several Arizona State University experts are on a team that created a new way to put a price on the rainforest in order to save it, and on Friday they won the top award in the prestigious $10 million XPRIZE Rainforest competition for their work.

Team Limelight Rainforest, previously called Team Waponi, which includes four ASU professors plus more than three dozen other scientists from around the world, invented a technology to measure biodiversity in the rainforest. 

They spent five years developing and refining their solution, which involves using a drone to drop devices onto the tree canopy to measure sound, record images, and take insect and plant samples.

The project combines artificial intelligence data analysis with Indigenous knowledge from the people in the region.

Garth Paine, an expert in acoustic ecology, created the bioacoustics recorders in the Limelight devices to analyze species density. He was thrilled with the XPRIZE victory, which was announced Nov. 5 at the G20 Social Summit in Rio de Janeiro, Brazil.

“It’s unbelievable, to be honest,” said Paine, who is a professor of digital sound and interactive media in the School of Arts, Media and Engineering and the School of Music, Dance and Theatre, as well as a senior Global Futures scientist.

“This is going to affect how the United Nations values and makes international agreements about the protection of the rainforest, and that's enormous. This is massively impactful.”

Team Limelight Rainforest was one of six finalist teams in the competition, taking home the grand prize of $5 million that will go toward further developing their solution.

The final round was held in the Amazon rainforest in Brazil in July, where each team deployed its solution for 24 hours and then had 48 hours to analyze the results. Team Limelight Rainforest identified over 250 different species and 700 unique plant and animal populations from the 24-hour deployment, the highest amount of biodiversity observed among finalist teams.

Their solution was judged to be the most scalable and easy to use for local communities and conservationists in the field.

Besides Paine, the other ASU members of Team Limelight Rainforest are:

• Pavan Turaga, professor and director of the School of Arts, Media and Engineering and professor in the School of Electrical, Computer and Energy Engineering.
• David Manuel-Navarrete, associate professor in the School of Sustainability and affiliate faculty in the Center for Behavior, Institutions and the Environment.
• Tod Swanson, associate professor of religious studies in the School of Historical, Philosophical and Religious Studies and a senior Global Futures scientist.
• Ankita Shukla, now an assistant professor at the University of Nevada, Reno, who started working on Team Limelight Rainforest while she was a postdoctoral research scholar in the School of Arts, Media and Engineering at ASU.
• Nicholas Pilarski, an associate professor in The Sidney Poitier New American Film School who creates immersive experiences, was involved in the earlier stages of the project.

The goal of the XPRIZE Rainforest competition was to find a way to measure the value of the rainforest by quantifying its biodiversity, Paine said.

“XPRIZE is interested in addressing this fundamental challenge, which is essentially, how do we protect the lungs of the Earth?” he said.

“But to date, we've had no mechanism to quantify that.

“And because the world works on money, there was no way of quantifying the value of those forests and therefore what they’re worth to the international community.”

If a value can be place on biodiversity, it will be easier for the international community to work with countries to leave biodiverse habitats untouched, improving the long-term health of the planet, instead of clearing the rainforest for a short-term benefit.

The Limelight Rainforest solution has several components, starting with a drone that delivers the sampling platform to the treetops. Then the platform collects bioacoustic data, images of insects that are attracted by a special light and insect specimens. The platform sends the data in real time to a base that identifies species using machine learning. Limelight reduces the time it takes to identify DNA from the environment by using a portable molecular lab kit. Then the drone removes the device.

One huge challenge to measuring the many kinds of animals and plants in the rainforest is that most of the species are still unknown. So the team had to find a way to measure the unknown.

Three years ago, Paine brought Turaga, a machine-learning expert, onto the project to help quantify the diversity of sound being recorded. Turaga began by training a model with existing datasets of rainforest bird sounds.

That failed when it was deployed in the field.

“So that was a year of work to train these bird-call, sound-classifying techniques. And then we were told that nothing worked,” Turaga said.

“So then we backtracked and the idea was to not classify everything as a specific type of bird, but to look at some aggregate measures of what we call feature-level diversity.”

The machine-learning team was able to identify different features of sounds into a kind of bird-call profile that was not necessarily specific to a bird species, and then measure all the different profiles.

“So it's a weaker version of diversity but it was something that we felt was more robust, and I think it worked,” Turaga said.

Complex problems require transdisciplinary solutions, and Team Limelight Rainforest includes a diverse array of experts in biology, neuroscience, ecology, bioinformatics, bird and insect identification, and anthropology.

Manuel-Navarrete and Swanson worked to ensure that Indigenous knowledge and perspectives from the Kichwa and Waorani groups were embraced in the project. Both professors work with Iyarina, an Indigenous-focused research and education organization in Ecuador. 

Swanson and his colleagues have been creating a series of videos about the Indigenous relationship to plants and animals that will be turned into a searchable platform.

Paine said that the work of Swanson and Manuel-Navarrete is a critical component of any attempt to place an economic value on the rainforest.

“If we talk to the Indigenous people that have lived there for centuries, they have a quite different perspective, and it’s about the forest as a living part of their family,” he said.

“It was really important to acknowledge that the forest is a lot more than just a collection of animals and microbes and so on.”

And Paine’s expertise as a musician brought a valuable perspective, Turaga said.

“That's not the type of insight you would find in the machine-learning circles or the engineering circles.

“It really came through this long track record of him spending time in these natural environments and really paying attention to the sounds he was hearing because he was trained in the music ways of thinking about sound.”

Paine, a flutist and composer, said he’s been exploring sound all his life.

“There’s a really powerful story to be told there, which is that the creative mind can and often does lead these projects and brings out innovation that wouldn't otherwise be there,” he said.

“The creative mind is what we breed here in Herberger, and I think it's the most valuable thing on the planet.”