The world’s rainforests are disappearing even before they are fully explored.

These ecosystems are thought to harbor tens of thousands of unknown vertebrates, trees and insects. Even for the species that have been found, there’s a dearth of information about how widespread they are, or what’s happening to their populations.

Yet the forests are falling faster than scientists can investigate this wellspring of biodiversity. Last year, 3.7 million hectares of tropical rainforest were logged or burned, the equivalent of 10 soccer fields every minute, according to the World Resources Institute.

While there’s no technological solution to the social, political and economic forces driving this devastation, some new inventions may help speed the cataloging of biodiversity in these forests, a critical step for conservation efforts.

Last week, the winners were announced in a $10 million competition put on by XPrize Rainforest to build devices that could survey 100 hectares of rainforest in a single day. The results are a testament to the growing conservation power of tools such as drones, DNA samplers, audio and video recorders and artificial intelligence, especially when paired with old-fashioned human ingenuity and the pressure to save these forests before it’s too late.

“These tools are ideally suited for implementation to monitor, manage and protect tropical rainforests globally,” said Peter Houlihan, executive vice president for biodiversity and conservation at XPrize.

The $5 million top prize went to a team whose drone-delivered tree-top labs identified 700 different types of animals and plants, including 250 distinct species over a 24-hour period (plus two days to analyze the data) while exploring a section of rainforest in the Brazilian Amazon. Developed by the U.S.-based nonprofit Limelight Rainforest under the leadership of Colorado Mesa University tropical ecologist Thomas Walla, the devices combine insect traps, audio recorders and cameras into a constellation of gadgets suspended in a lightweight lattice of rods. A drone uses a robotic hand to gently set one of these little laboratories into the top of a tree. Illuminated at night to attract insects, the constellation of devices resemble small lanterns floating in an ocean of trees.

“Limelight has the potential to revolutionize the rate at which we monitor and assess biodiversity with technology that’s small enough to fit in a backpack,” said Walla. “Compared to existing technologies that may sequence 1,500 species over 2 years, Limelight has the potential to sequence 2,500 in 1 week.”

Second place and $2 million went to a Yale University-based team that developed a fleet of semi-autonomous flying drones that could explore the jungle, gathering images, sounds and environmental DNA. The approach relies partly on sophisticated AI programs that can use satellite images to predict what species will be in a place before the drones are deployed, as well as to scan audio and images collected by the drones to automatically identify species or flag unusual ones for a person to check. During the final competition, the team was able to identify 225 different species during a 24-hour survey.

A team made up chiefly of Brazilian researchers, meanwhile, vacuumed up data using a dizzying array of tools. Small, wheeled robots collected water samples, bits of leaves on the ground and insects, especially blood sucking ones such as mosquitos whose meals could be studied for the DNA of the animals on which they fed. Flying drones recorded high-definition images of the canopy, while acoustic recorders captured the sounds of insects, birds, bats and amphibians. Their approach took home the third place prize of $500,000.

Despite all the remarkable gadgetry, there are limitations for the new tools. For instance, if a species and its genetic fingerprint haven’t already been catalogued, a drone collecting DNA from the forest won’t know what it’s found.

Pedro Brancalion, a scientist at the University of São Paulo who helped lead the Brazilian team, said their winnings would be used to help tackle this problem. “The main challenge we have today is to create the libraries, to create the data collection of DNA and other types of biological material to enable us to use the technologies, to upscale the monitoring of tropical biodiversity.”

Even with all this automation, there’s still more than enough work to be done by scientists going out in the jungle.

Photo courtesy of Cat Kutz