October 1, 2020

Patterns in blue whale songs show whether the whales are feeding or migrating

Stanford graduate student William Oestreich watches the blow from a blue whale in Monterey Bay. Image: Jeremy Goldbogen

After analyzing five years of blue whale calls, researchers at MBARI, Stanford University, and several other institutions discovered that the time of day that the whales sing changes dramatically depending on whether the whales are feeding or migrating. This discovery will give scientists new insights into the lives of these endangered giants of the sea and might eventually be used to help protect them from human threats such as ship strikes.

It turns out that when male blue whales are swimming along the coast as part of their annual migration, they are most likely to sing during the daytime. But when they are staying in one area and feeding on krill, they mostly sing at night. The researchers made this discovery by analyzing whale calls recorded by MBARI’s deep-sea hydrophone and comparing these with recordings and other data from temporary tags attached to living whales.

Blue whales are the largest animals on Earth. They are also among the loudest, making vocalizations that can travel for hundreds of kilometers underwater. “Sound is a vital mode of communication in the ocean environment, especially over long distances,” explained William Oestreich, a graduate student in biology at Stanford University’s Hopkins Marine Station. “Light, or any sort of visual cue, is often not as effective in the ocean as it is on land. So many marine organisms use sound for a variety of purposes, including communicating and targeting food through echolocation.”

Although whale songs have been studied for decades, researchers have had limited success in deciphering their meaning. By recording songs from groups of whales (using hydrophone data), and from individual whales (using tags on the animals themselves), the researchers were able to detect patterns in the trills and bellows of blue whales. These patterns indicate when the whales are migrating from their feeding grounds off the North American coast to their breeding grounds off Mexico and Central America—a 6,400-kilometer (4,000-mile) annual journey that ranks among the longest in the world.

“We decided to compare daytime and nighttime song patterns from month to month, and there, in the divergence and convergence of two lines, was this beautiful signal that neither of us really expected,” said John Ryan, a biological oceanographer at MBARI and senior author of the paper. “As soon as that image popped up on the screen, Will and I were both like, ‘Hello, behavior.’”

This graph, based on data from a tagged blue whale, shows how the whale made feeding dives during the day, but stayed near the surface and sang at night. Image courtesy of William Oestreich

The researchers described their findings in a new paper in the journal Current Biology. Coauthors on the paper include faculty and students at Stanford as well as researchers from Cascadia Research Collective, the University of California Santa Cruz, the U.S. Naval Postgraduate School, Southall Environmental Associates, and Moss Landing Marine Laboratories.

Supping and singing

To capture whales singing, the researchers used two advanced recording technologies: an underwater microphone (hydrophone) and tags that the researchers placed on individual whales to record their calling, foraging, and migratory behaviors.

In 2015, MBARI deployed a hydrophone on the seafloor just outside Monterey Bay, 900 meters (3,000 feet) below the surface. The hydrophone is wired to MBARI’s MARS undersea cabled observatory, which provides power and communications to instruments on the deep seafloor. A live audio stream from the hydrophone is available on MBARI’s web site.

After analyzing five years of hydrophone recordings and focusing on the sound frequencies most used by whales, the researchers noticed distinct seasonal changes. During summer and early fall, the whale arias grew louder and were sung mostly at night, typically peaking in October or November. But soon after this annual peak, singing became more of a daytime activity. This change coincided with the time of year when the whales stopped feeding and began their annual migration to the warmer waters off Mexico and Central America .

Although the existence of daytime/nighttime differences in singing behavior had been noted in previous research, the seasonal change in day/night balance had not. The whale-borne tags, developed in the lab of Stanford biologist Jeremy Goldbogen (another senior author on the paper) helped explain the meaning of these 24-hour patterns. Fifteen tags carried accelerometers that recorded not only the motions of the tagged animals, but also their songs (in the form of vibrations within the animals’ bodies). Some tags also carried miniature hydrophones.

Researchers place a temporary tag on a blue whale to record its movement and vocalizations. Image taken by James Fahlbusch of Stanford University under NMFS research permit 1611.

During the summer, the accelerometers showed that the tagged whales spent much of their day feasting, bulking up for their long journey. At night they fed less often and spent more time singing. Later in the year, when GPS tracking showed that the whales were beginning to migrate, the animals sang most frequently during the daytime.

“In the hydrophone data we saw really strong patterns over this enormous spatial domain. When we saw the exact same pattern on individual animals, we realized that what we’d been measuring over hundreds of kilometers is actually a real behavioral signal—and one that represents the behavior of many different whales,” said Oestreich. “As an ecologist, it’s very exciting to observe so many whales, simultaneously, using one instrument.”

Listening and learning

This research lays the groundwork for possibly predicting blue whale migration based on the transitions between the different song schedules. Such forecasts could be used to warn ships’ crews to watch out for whales or take precautions to avoid hitting them. The researchers also hope that further analysis of the acoustic data will reveal more about whale behavior in response to environmental changes, such as warming waters and fickle food supplies.

“If we can detect differences in migration and foraging in response to changes in the environment, that is a really powerful and important way to keep an eye on this critically endangered species,” said Goldbogen. “That’s economically important, ecologically important, and also culturally important.”

Oestreich is currently pursuing a related question: If we can use this signal to determine whether whales are foraging or migrating, could the whales be using it that way too? It’s possible, said Oestreich, that a lone whale might listen around before giving up on feeding and heading south.

“Blue whales exist at incredibly low densities with enormous distances between them, but are sharing information in some way,” said Oestreich. “Trying to understand that information sharing is one motivation, but also potentially using that signaling as a means to study them is another exciting possibility.”

Article adapted from a Stanford University news release


Original journal article:
Oestreich, W.K., J.A. Fahlbusch, D.E. Cade, J. Calambokidis, T. Margolina, J. Joseph, A.S. Friedlaender, M.F. McKenna, A.K. Stimpert, B.L. Southall, J.A. Goldbogen, J.P. Ryan Animal-borne metrics enable acoustic detection of blue whale migrationCurrent Biology. October 1, 2020. doi 10.1016/j.cub.2020.08.105


The video below shows a spectrogram of blue whale “A” “B” and “C” calls, paired with audio of these same calls played back at ten times their original speed to make them easier to hear. The audio was recorded from MBARI’s deep-sea hydrophone on the seafloor just outside of Monterey Bay. New research shows that the day/night pattern of “B calls” can be used as an indicator of whether the whales are feeding or migrating.

For additional information or images relating to this article, please contact: Kim Fulton-Bennett
831-775-1835, kfb@mbari.org

MBARI Researchers

John Ryan

Senior Research Specialist

Craig Dawe

Technical Support Manager/ MARS Manager