Can the Cloud Save Endangered Species from Extinction?

If nature is cruel then humans are worse.

So suggests the World Wide Fund For Nature (WWF), which says at least 10,000 plant and animal species – and maybe even as many as 100,000 – go extinct every single year. The United Nations likewise says nearly 1 million species are now in danger of going extinct, which is more than at any other point in human history.

Humans are to blame. Through degradation, fragmentation, overfishing, overharvesting and overhunting, they’re permanently fraying fragile ecosystems in which all species rely on one another for their mutual survival, according to Antarctic ecologist Michael Wethington, who said the key to species’ long-term survival is biodiversity.

“If you lose any component from the local ecosystem, then you have cascading effects down the line – not just in this moment, but for years to come,” Wethington said.

When they deplete the Earth’s natural biodiversity, humans aren’t just harming flora and fauna. They’re also harming themselves, points out Oswald Schmitz, professor of population and community ecology at Yale University, who says biodiversity supports vital industries on which humanity relies for food and shelter.

“We need biodiversity just for basic necessities of human living – to put groceries on the table or to have building supplies for housing,” Schmitz said. 

“Biodiversity provides commodities that we use. You can’t do forestry or agriculture without having healthy soils or having a diversity of pollinator species that pollinate your plants. Keeping the functional structure of ecosystems intact can actually minimize a lot of harm that can come and hit us.”

That harm includes age-old threats like famine, drought and disease, all of which biodiversity can help mitigate. 

“Biodiversity provides services that support our wellbeing and our ability to live because they engage in functions that drive processes like air purification, water purification and nutrient cycling,” said Schmitz.

Fortunately, endangered plants and animals are beginning to attract some much-needed reinforcements from a surprising place: cloud computing.

The Internet of Things (IoT), artificial intelligence (AI), machine learning and drones are just a few of the cloud-based technologies that conservationists are using to monitor and track endangered species, and to wage war against the developers and poachers who threaten them.

Surveying Survival

Saving species starts with surveying them, according to Wethington, who says technology is making it easier to accurately monitor plant and animal populations. In particular, drones, imagery from which researchers can use to count endangered animal populations and ensure the continued health of at-risk species.

“Trying to survey species is logistically very challenging, so we are able to use drones to survey entire animal colonies, take thousands of images and bring that back home,” he said. 

Drones already are monitoring penguin populations, scanning seabird colonies and tracking black bears.

“There’s near-endless potential and application for drone usage within wildlife science,” Wethington continued. “Drones are great for getting data in a short amount of time and validating other methods that we are using. They fly low, so you don’t have issues with cloud cover, which is a concern that you have with satellites.”

Plus, drones can go places people can’t – or shouldn’t. 

“Some environments are very treacherous and hard to access, so drones enable us to work around those constraints without having to assume the risk of dangerous ice climbing or technical mountaineering,” Wethington said. 

“In addition to that, drones can develop three-dimensional surface models, which are necessary if you’re doing, say, a hydrological analysis.”

When they know with confidence a specie’s numbers, researchers can quickly spot when the population is dwindling, then sound the alarm and intervene when necessary, according to Wethington, who said AI and machine learning also can help with surveying.

“AI and deep learning are critical components of doing this work at scale,” he explained. “Making the most of the data is impossible without AI and machine learning. We use a machine learning algorithm that can automatically count heads in aerial imagery and empirically identify these things in a matter of minutes, versus 60 to 80 hours counting heads manually.”

Nonprofits like Wild Me are making machine learning more accessible to everyday conservationists by training algorithms that can identify dozens of different species, from sea turtles to giraffes. Some of its algorithms can even identify individual animals based on their unique markings.

Wild Me’s open-source platform, Wildbook, resides on the cloud so that others can easily access it and contribute to the organization’s growing virtual catalog of wildlife imagery.

Charitable organizations also are pledging to the cause. Among them, for example, are the Gordon and Betty Moore Foundation – started by Intel co-founder Gordon Moore, it recently committed $5 billion to help conserve 30% of the planet – and Google Cloud, which has partnered with the Zoological Society of London to test a machine learning algorithm that can recognize gunshots, allowing conservationists to pinpoint locations where extra law enforcement might be needed to stop poachers.

The Elephant Listening Project also uses AI and acoustic monitoring for conservation. Utilizing IoT sensors deep inside the African jungle, it monitors soundscapes in hopes of capturing calls from the critically endangered forest elephant. Its machine learning algorithms sift through terabytes of audio data in the cloud in order to isolate the elephants’ unique voices in the noisy wilderness, which allows researchers to study the species’ declining population and protect it from poachers who covet elephants’ prized ivory tusks.

The Hunter Becomes the Hunted

Poaching has been illegal in at least some forms and in some places for centuries. And yet, it remains a catastrophic problem for wildlife, sometimes putting entire species on the path towards extinction.

Like the Elephant Listening Project, Wildlife Protection Solutions is using technology to stop poachers in their tracks. Whereas the former uses acoustic sensors, however, the latter uses cameras, images from which it analyzes with AI. When its algorithms detect potential threats, it can deploy rangers in order to stop wildlife crime before it happens.

“We found that cameras were a very scalable way to gather information from the field,” said Wildlife Protection Solutions Executive Director Eric Schmidt. 

“They’re great for collecting information, but they’re not so good at being able to distinguish what matters and what doesn’t. So we pair the camera inputs with a variety of different AI algorithms that can search for things like humans where they shouldn’t be, allowing us to sift through the inordinate amount of data that the cameras send through.”

Thanks to machine learning, the system improves over time, using heaps of imagery to hone its ability to recognize animals and threats. 

“The whole concept is largely not possible without machine learning and AI – there’s just too much data,” explained Schmidt, who said his company processes nearly 10,000 images a day. 

“The reliability of a human to sift through those and accurately make those calls on a 24/7 basis is just economically not possible and physically highly unlikely.”

Compared to humans, technology can spot more bad actors, faster. That’s a powerful new advantage in the fight against poachers – and it could become even more powerful with the help of technologists at Harvard University, who are building AI software that can predict poacher’s actions before they even happen.

“Where are the poachers going to hit next? There are thousands of square kilometers in national parks. There are hundreds of rangers. They can’t be everywhere. If we can tell them where they need to be to stop the poaching, that’s important,” Milind Tambe, director of Harvard’s Center for Research in Computation and Society, told Hewlett Packard Enterprise’s enterprise.nxt.

Cloud-Based Conservation

Wherever wildlife and poaching data come from – sensors, cameras, microphones or drones – cloud computing is what ultimately makes it actionable, allowing everyone from researchers to rangers to glean critical and timely insights that can save the life of an endangered animal.

“Our entire infrastructure is cloud based,” Schmidt said. “It’s quite easy and dramatically reduces the amount of time and energy we have to dedicate to investing in IT infrastructure and scalability concerns. As we continue to grow the camera network, we can simply migrate to higher-end and better cloud resources without having to worry about hardware, data centers, additional storage and all the nightmares that come along with rapid growth.”

By protecting endangered plants and animals and preserving biodiversity, the cloud is doing its part to make life on Earth once more beautiful and more sustainable.

“It is horrifying to think about the possibility that we may be leaving a world behind where keystone species like tigers, elephants and rhinos may just be gone,” concluded Tambe, who said technology and biodiversity are the keys to leaving behind a healthy planet for the next generation. 

“We don’t want to have to tell our children, ‘Well, they’re all gone.’ We don’t want a world like that.”

Chase Guttman is a technology writer. He’s also an award-winning travel photographer, Emmy-winning drone cinematographer, author, lecturer and instructor. His book, The Handbook of Drone Photography, was one of the first written on the topic and received critical acclaim. Find him at chaseguttman.com or @chaseguttman.

© 2022 Nutanix, Inc. All rights reserved. For additional legal information, please go here.