Deep beneath the ocean’s surface, sperm whales engage in complex acoustic conversations, their clicks echoing for kilometres through the dark waters.
Now, a groundbreaking scientific endeavour is allowing researchers to eavesdrop on these exchanges in real-time, thanks to an autonomous underwater robot.
Sperm whales utilise distinct clicks for navigation and hunting, alongside patterned sequences known as “codas”, believed to be crucial for communication.
While their vocal prowess was first identified in 1957, understanding these marine giants has remained a significant challenge. They routinely dive to depths exceeding 1.6 kilometres for approximately 50 minutes each hour, making continuous observation incredibly difficult.
However, scientists are overcoming this hurdle with an innovative underwater glider.
This small robot, which subtly alters its buoyancy to ascend or descend, is equipped with four hydrophones and a “backseat driver” feature designed to track the whales by their voices.
David Gruber, founder and CEO of Project CETI, and a co-author of the study published this week in Scientific Reports, explained the process.
“The underwater glider is listening for whales via four hydrophones and then steering itself toward them using a feature called backseat driver,” Mr Gruber, also a professor of biology and environmental sciences at Baruch College at the City University of New York, stated.
“When the glider detects the distinctive vocalisations of sperm whales, the software on board identifies where that sound is coming from and automatically communicates with the glider’s navigation system to change directions and follow the whale.”
“You can think of it as a quiet, long-distance explorer, more like a soaring albatross than a motorised vehicle, steadily travelling through the ocean while listening and gathering information as it goes,” Gruber said.
Traditional tracking methods rely on suction tags that fall off after a few days or on stationary sensors that lose contact when whales move away. Project CETI also deploys hydrophones – underwater devices that detect and record sounds – towed from boats.
What makes the new robotic system different, Gruber said, is that it “can make decisions in real time while it’s still underwater,” rather than recording acoustic data for later analysis.
Previous methods allowed scientists to reconstruct where a whale had been, but not actively follow it at the moment. The new approach “continuously updates the glider’s path so it can stay with a single whale for extended periods – potentially months,” Gruber said.
The ability to track whales for longer periods marks what Gruber called a shift “from brief encounters to continuous relationships,” allowing scientists to stay with the same whale or group instead of relying on short, opportunistic glimpses and to see patterns in how whales coordinate, socialise and respond to their environment over time.
Such data could also help answer longstanding questions about how sperm whales communicate.
“By following mother-calf pairs over time, we can begin to see how calves pick up vocal patterns from their mothers,” Gruber said.
The system could also reveal how whales react to human activity, allowing researchers to track the way their communication changes in the presence of human-made noise and offering a clearer picture of how shipping, offshore construction or fishing affect them.
By linking whale behaviour with environmental pressures, the technology could inform more precise, evidence-based policy decisions such as when to reduce ship speeds, reroute traffic or implement fishing restrictions to minimise disruption in sensitive areas, the researchers said.
Developing the system “brings us closer to understanding another form of intelligence on Earth, which has implications not just for conservation, but for how we think about communication and life beyond our own species,” Gruber added.
Yet, precise localisation remains a challenge, as the glider can detect the direction of a whale but not its exact position, limiting its ability to distinguish between individual whales and track them accurately.
Communication is another constraint. The robot must surface every few hours to send and receive updates, making long-term, real-world monitoring less seamless.
For Gruber, the moment the glider acted on its own offered the first real glimpse of what this technology could become.
“We’re beginning to build systems that can operate independently and respond to the natural world as it unfolds,” Gruber said.