AI powered listening posts are now protecting endangered elephants from poachers over an area of 580 square miles in Nouabalé-Ndoki National Park in the Republic of Congo.
It means scientists can learn critical information about the elephants’ habits and patterns in a fifth of the time they used to so park managers can be alerted about potential risks before it’s too late.
They are doing this by creating sensors that can distinguish low-frequency, long-duration elephant calls from other rainforest sounds.
Before it took up to eight weeks to decipher the sound data but now these samples can be analysed in 22 days, and experts hope this could soon be done even faster.
Forest elephants are a distinct species from the more populous – although also endangered – savanna elephants.
Their vast range and the thick rainforest canopy make them extremely difficult to track by land or air.
The number of forest elephants in central Africa has plummeted from an estimated 100,000 in 2011 to fewer than 40,000 today.
AI powered listening posts are now protecting endangered elephants from poachers over an area of 580 square miles in Nouabalé-Ndoki National Park in the Republic of Congo
Scientists from Cornell University currently track elephants with acoustic sensors as part of their Elephant Listening Project (ELP).
However, the forests are so remote and the sound files so huge it takes months to collect and analyse the data – too long to rescue the animals from poachers or other threats.
For example in 2014, scientists learned 25,000 forest elephants had been slaughtered by ivory poachers in Gabon’s Minkébé National Park.
Before anyone realised those elephants were in danger, they’d been wiped out.
Fifty sensors have now been installed throughout a 580-square-mile area of Nouabalé-Ndoki National Park through a collaboration with an AI startup called Conservation Metrics.
They generate seven terabytes of data every three months – the equivalent of two million iTunes songs.
New developments in machine learning and deep neural networks make it possible to analyse these enormous files faster and with much higher accuracy.
This means scientists can alert park managers about red flags in less time.
Fifty sensors have been installed throughout a 580-square-mile area of Nouabalé-Ndoki National Park. They generate seven terabytes of data every three months – the equivalent of two million iTunes songs
‘A key thing this collaboration will do is speed things up, so we can show the people who manage the national park that we can provide information that will make a difference,’ said Peter Wrege, director of the Elephant Listening Project, part of the Cornell Lab of Ornithology.
‘If it takes us a year to figure out what elephants are doing in the forest, it’s already too late.’
Using the data from acoustics, researchers can create maps showing the elephants’ habits.
For example, a recent map revealed large numbers of elephants congregating in an area adjacent to a logging site and close to roads, making them highly vulnerable to poachers.
‘What the Elephant Listening Project is doing in terms of working with collaborators on these sites in Africa is really impressive, but the logistics are really hard,’ said Matthew McKown, CEO of Conservation Metrics, which recently received a two-year Microsoft AI for Earth grant for this work.
‘It’s a truly ambitious project, and it’s the first time we’re actually realising the potential of these automated monitoring approaches.’
New developments in machine learning and deep neural networks make it possible to analyse these enormous files faster and with much higher accuracy
New research has proven that elephants’ emotional characteristics are similar to those of humans.
It turns out the animals have distinct personalities.
They can be aggressive, attentive and outgoing.
For the study scientists asked elephant riders, or mahouts, to answer questions about the behaviors of the animals they worked with each day.
A new study has found that elephants, like humans, have distinct personalities. They can be aggressive, attentive and outgoing. Pictured is an elephant with its mahout, or rider, who the animal works with each day in Myanmar’s timber industry
Dr Martin Steltmann, who worked on the new report, explained how his team defined the traits that categorize elephants.
He said: ‘Attentiveness is related to how an elephant acts in and perceives its environment.
‘Sociability describes how an elephant seeks closeness to other elephants and humans and how popular they are as social partners.
‘Aggressiveness shows how aggressively an elephant acts towards other elephants and how much it interferes in their social interaction.’
Dr Steltmann’s team is hopeful the new research can aid in elephant conservation efforts.
Part of the grant from Microsoft includes access to Azure, its powerful cloud server.
Once the team develops compatible software it could potentially complete the analysis using Azure in a single day.
Eventually, ELP and Conservation Metrics hope to create a way to run the artificial intelligence tools directly in the African forests; then the numbers of elephants in each area could simply be texted to the United States.
‘It’s years away, but if you look at the technology that is being developed for commercial purposes, it seems feasible, whereas a few years ago we would have just said it was a dream,’ Mr McKown said.
Mr Wrege said his hope is that acoustic monitoring can become fast and effective enough for park managers in Africa to appreciate its value.
‘Acoustics isn’t going to stop the poaching, but I do think it offers maybe the only way we can get information regularly enough,’ he said.
‘It’s daunting, but it’s worth it, and it can be done. We just have to keep at it.’