Undersea "Caller ID" Can Pinpoint Exactly Which Whale Is Calling
By tagging individual whales with audio sensors, researchers have been able to track their calls more accurately than ever before.
Scientists from Syracuse University, Aarhus University, the Woods Hole Oceanographic Institution, the Center for Coastal Studies, the Whale Center of New England, the University of California at Santa Cruz, and Stellwagen Bank National Marine Sanctuary have built "caller ID" with a difference: it identifies individual whales by their sound.
“By simultaneously tagging all whales in a group, we were able to compare how loudly calls were recorded across tags to infer who was calling," explains first author Julia Zeh of the team's work. "This in turn lets us look at individual and group-level communication in ways that we couldn't before."
"This information can give us insight into how whales coordinate behaviors," Zeh continues, "how their calls relate to what they’re doing, what types of calls they use and what information they might exchange in group communication. Understanding acoustic sequences within and between individuals also gives us insight into the complexity of the humpback whale communication system."
Tracking whales through audio recordings isn't a new concept, but prior approaches have had a problem: by using undersea microphones to record all sound in a given area, it's difficult to figure out which of a given group of whales is "speaking" at any given time. To solve this, the researchers deployed acoustic "tags" which could be applied to individual whales — accurately identifying their audio versus those of others in the area.
"Having information from tag data about call rates and timing can improve [population] count estimates," Zeh adds of the benefit to the approach, which has not previously been robustly proven for humpback and other large baleen whales. "For example, having 10 calls doesn’t necessarily mean there are 10 whales, but potentially two whales calling back and forth, or one whale producing sequential calls."
The team's work has been published in the Royal Society Open Science journal under open-access terms.