In February 2020, behavioral ecologist Mickey Pardo observed an intriguing scene in Kenya’s Samburu National Reserve when a female elephant named Matisse communicated with her newborn using deep vocalizations called “rumbles.” A new study published in Scientific Reports has used machine learning to analyze these rumbles, revealing different types such as “coo,” “contact,” “let’s-go,” “cadenced,” and “greeting” rumbles. The study found that these vocalizations, which are low and growl-like, may have similar acoustic properties in similar situations, suggesting a form of communication. The phenomenon of “vocal convergence,” where rumbles become similar during group coordination, was also noted, hinting at possible agreement among elephants. This research highlights the importance of combining computational analysis with real-world observations to better understand animal behavior.
QUESTION: How might understanding elephant communication change the way we approach wildlife conservation?
