Beluga whales and narwhals are closely related and seasonally inhabit the same Arctic waters. Distinguishing between their different vocalizations enables effective passive acoustic monitoring to study each species’ populations and behaviors.

Using recordings collected in the Arctic, Zahn et al. proved one-third octave levels (TOLs) are an effective measure to differentiate between the two whale’s echolocation calls.

To obtain recordings of the whale vocalizations, the team deployed ocean moorings near glaciers in Melville Bay in Northwest Greenland.

“Narwhals are known to visit glacier fronts during the summer months, so these data provide an important time series of narwhal presence/absence in their summering grounds,” said author Marie Zahn. “Collecting any in situ data near calving glaciers in the Arctic is extremely risky since huge icebergs can damage equipment. In fact, one of the three moorings deployed from our study was destroyed by an iceberg. Therefore, the recordings we used represent a unique dataset that is difficult to obtain.”

The team used TOLs to identify which animal produced echolocation signals in their recordings.

“In general terms, TOLs represent the total sound energy in decibels for a specific frequency range with upper and lower limits defined by one-third octaves,” said Zahn. “In the field of bioacoustics, we quantify variations in sound energy across a frequency bandwidth for a particular sound.”

The difference in energy between the 16-25 kHz and 25-40 kHz TOLs successfully classified the narwhal and beluga vocalizations.

Their technique can be automated and applied to other recordings or fitted to classify different whales. As global warming impacts the beluga and narwhal habitat, effectively monitoring their population and behaviors will be crucial to informing any interventions.

Source: “Accurate species classification of Arctic toothed whale echolocation clicks using one-third octave ratios,” by Marie J. Zahn, Michael Ladegaard, Malene Simon, Kathleen M. Stafford, Taiki Sakai, and Kristin L. Laidre, Journal of the Acoustical Society of America (2024). The article can be accessed at