A New Type of Lion Roar Was Just Discovered by Scientists

There is a lot more to the lion’s roar than the opening credits of an MGM film. And now, scientists have even discovered a previously unrecognized type of lion roar. It is hoped that this discovery will aid wildlife monitoring and conservation efforts in the future. Read on to find out more.

Why Study the Lion’s Roar?

The African lion (Panthera leo) is a very vocal creature! Their repertoire of sounds is extensive and includes mews, snarls, chuffs, grunts, moans, and, of course, roars. The roar is one of the sounds they emit quite regularly, and it is used for long-distance communication because it carries a long way. Lions use their roar to communicate with other pride members and to advertise their territory. Both males and females roar, and they are affected by age. Lions under 2.5 years and nomadic males who do not have an established territory are less likely to roar. Interestingly, lions also roar most often in the hours leading up to dawn and when they are close to water. There are risks associated with roaring – the lion is advertising their location!

Because each lion’s full-throated roar is unique, researchers have used acoustic recording devices to monitor populations. However, achieving accurate identification has been challenging.

Defining and Recording a Roar

If scientists are going to use the roar to monitor lion populations, there needs to be a clear and consistent definition of a full-throated roar. Using human perception of this alone has not proved to be accurate enough. To address this, researchers have used spectrograms—a visual representation of sound intensity—to show that a roaring bout consists of different stages. Researchers from the University of Exeter, in collaboration with other partners, set out to analyze the different components of a lion’s roaring bout. This study is significant because it was the first time that artificial intelligence was used to automatically differentiate between lion roars.

What the Study Shows

The study was conducted at the Matambwe sector of Nyerere National Park in southern Tanzania. Advanced machine learning techniques and an automated data-driven approach were used to classify recordings of full-throated roars. Artificial intelligence was able to detect subtle differences in the roars that the human ear could not distinguish.

Importantly, the study found that a lion’s roar includes both the familiar full-throated roar and a newly identified intermediary roar. The study also improved the identification of individual lions. Speaking in this press release by the University of Exeter, lead author Jonathon Growcott said, “We believe there needs to be a paradigm shift in wildlife monitoring and a large-scale change to using passive acoustic techniques. As bioacoustics improve, they’ll be vital for the effective conservation of lions and other threatened species.”