A biologist named Jenna Lawson has hidden 350 audio monitors in trees across Costa Rica’s Osa Peninsula to spy on the elusive and endangered Geoffrey’s spider monkeys that dwell high in the rainforest canopy. The devices, placed in the lush region, recorded the sounds of the forest and surrounding countryside for a week, collecting an overwhelming amount of data that Lawson could have spent years listening to. Instead, she utilized artificial intelligence systems trained to recognize spider monkey calls instantly and detect their movements. This innovative approach led to troubling findings about the health of the wildlife refuge in Costa Rica, shedding light on the urgent need for AI-assisted wildlife surveillance as 28% of all plant and animal species are at risk of extinction.
In an effort to tackle technical challenges related to AI-assisted wildlife surveillance, Microsoft’s philanthropic AI for Good Lab has introduced a new hardware and computing system designed for eavesdropping on remote and wild areas around the world. Referred to as Sparrow, short for Solar-Powered Acoustic and Remote Recording Observation Watch, these devices are powered by the sun and energy-efficient AI computer chips, enabling them to operate for longer periods without human intervention. They can transmit data online via low-Earth orbit satellites, making them ideal for monitoring biodiversity hotspots. The initiative aims to set up Sparrow devices in various locations globally, including remote areas like the Amazon rainforest and the gorilla habitats in the Democratic Republic of the Congo, to aid in conservation efforts.
The bio-acoustic study conducted by Jenna Lawson highlighted the critical situation faced by spider monkeys in Costa Rica due to encroachments on their habitat. Placing recording devices outside Corcovado National Park, Lawson aimed to understand the impact of human activities on the monkey population. Compared to other primate species commonly seen or heard in Costa Rica, spider monkeys are more sensitive to human presence and changes in their environment, making them a key indicator species for ecological studies. Lawson’s findings revealed that the spider monkeys avoided paved roads and plantations, indicating that wildlife corridors meant to extend their range were not functioning effectively. This information was shared with local officials in an effort to improve conservation efforts in the region.
The use of unattended acoustic detection of animal sounds has proven valuable in various ecosystems, as highlighted in a recent Science paper. This technology can be beneficial for a wide range of applications, such as aiding sailors in avoiding collisions with large baleen whales in shipping channels. While there are challenges to overcome, such as humidity affecting jungle monitors and wildlife unintentionally interfering with the devices, the potential of using audio monitors to study wildlife behavior at a larger scale and lower cost holds promise. By reducing human influence on animal behavior and providing a more accurate account of their movements, researchers can gain valuable insights into the ecological dynamics of different species, ultimately aiding in conservation efforts.
Overall, the advancements in AI-assisted wildlife surveillance offer a promising approach to monitoring and studying endangered species in their natural habitats. By combining artificial intelligence with acoustic monitoring technology, researchers can gather valuable data on wildlife populations, behavior, and habitat use, enabling more efficient and accurate conservation efforts. The collaboration between tech giants like Microsoft and academic researchers demonstrates the potential for innovation in wildlife conservation, highlighting the importance of leveraging technology to protect biodiversity and address the urgent threats facing many plant and animal species worldwide.
