How Do You Count Endangered Species? Look to the Stars
The conversation startedover a fence dividing two backyards. On one side, an ecologist remarked thatsurveying animals is a pain. His neighbor, an astronomer, said he could see objectsin space billions of light years away.
And so began an unusualpartnership to adapt tools originally developed to detect stars in the sky tomonitor animals on the ground.
The neighbours, StevenLongmore, the astronomer, and Serge Wich, the ecologist, both of Liverpool’sJohn Moores University in England, made their backyard banter a reality thatmay contribute to conservation and the fight against poaching.
The scientists developed asystem of drones and special cameras that can record rare and e
ndangered species on the ground, day or night. Computer-vision andmachine-learning techniques that help researchers study the universe’s oldestand most distant galaxies can now be used to find animals in video footage.
Claire Burke, anastrophysicist at the university now leading the project, presented the team’slatest findings at the European Week of Astronomyand Space Science.
Keeping track of elusiveanimals, especially those that are endangered, isn’t trivial. First, it takestime and money to conduct manual counts on the ground or to shoot photos fromplanes in the sky. With video, cheaper drones and software, identifying animalshas become more efficient.
But cameras made fordaylight can miss animals or poachers moving through vegetation, and thedevices don’t work at night. Infrared cameras can help: Dr. Wich had been usingthem for decades to study orangutans.These cameras yield large amounts offootage that can’t be analyzed fast enough. So what do animals and stars havein common? They both emit heat. And much like stars, every species has arecognizable thermal footprint.
“They look like reallybright, shining objects in the infrared footage,” said Dr. Burke. So the softwareused to find stars and galaxies in space can be used to seek out thermal footprintsand the animals that produce them.
To build up a reference libraryof different animals in various environments, the team is working with a safaripark and zoo to film and photograph animals. With these thermal images they’llneed thousands – they’ll be able to calibrate algorithms to identify targetspecies in the ecosystems around the world.
The experts started withcows and humans in England. On a sunny, summer day in the year 2015, the teamflew their drones over a farm to see if their machine-learning algorithms couldlocate the animals in infrared footage. For the most part, they could.
But accuracy wascompromised when drones flew too high, cows huddled together, or roads androcks heated up in the sun. In a later test, the machines occasionally mistookhot rocks for students pretending to be poachers hiding in the bush.
Last September, thescientists honed their tools in the first field test in South Africa. here,they found five Riverine rabbits in a relatively small area. These shy creaturesare among the world’s most endangered mammals. Only a thousand have ever beenspotted by people.
The tests helped thescientists calculate an optimal height to fly the drones. The team also learnedthat animals change shape in real time (rocks don’t) as drones fly over. Andthe researchers found that rain, humidity and other environmental, atmosphericand weather conditions can interfere with proper imaging.
Scientists are refiningtheir system to account for these issues. In two years, Dr. Burke said, theyplan to have a fully automatic prototype ready for testing. Within five years,she hopes to sell systems at cost – today it is just around $15,000.
In the meantime, theseastro-ecologists are also working with search and rescue groups to help findpeople lost at sea or in fog. And starting in May, they will collaborate with conservationgroups and other universities to look for orangutans and spider monkeys in thedense forests of Malaysia and Mexico, as well as river dolphins in Brazil’smurky Amazon River. A&W