Sixteen hours of nonstop singing would wreck a human throat, a human diaphragm and probably a human marriage. A male humpback whale hanging in the warm water off Maui treats it as a normal shift.

That endurance comes from a paper published in Current Biology, by Julia Zeh, then a doctoral researcher at Syracuse University, and her co-authors. But the finding that has bioacousticians talking is smaller and stranger than stamina. Song is welded to the dive, phase by phase.

What a tablet-sized barnacle heard

Between 2018 and 2024 the team stuck suction-cup recording tags, each roughly the size of a tablet, onto the backs of 16 male humpbacks on their breeding grounds off Maui. As described by Syracuse University, the tags went on from small boats using long poles and, in later seasons, drones. They recorded sound and body movement at the same time, then lost suction, floated up and waited to be retrieved.

Recording both at once is the whole trick. Most whale song research has been done with hydrophones listening from a distance, which tells you beautifully what an animal sounded like and almost nothing about what its body was doing while it sang.

Descent, bottom, ascent

A singing humpback’s dive has a shape. Down, a long, steady stretch at depth, back up, then a breath at the surface.

Song has a shape too. Brief sounds called units combine into phrases, repeated phrases build a theme, and themes in sequence make a song.

Zeh and her colleagues found the two shapes lock together, as documented in the project’s public data archive. Certain themes showed up reliably on the way down, others on the way up, and a wider mix while the animal coasted along the bottom of the dive. Roughly one dive, one song.

Descent and ascent themes stayed consistent across whales and years, a strong hint that something other than personal taste is choosing them.

The lazy part of the dive is the showy part

So why would the deepest, stillest stretch carry the most variety?

The explanation offered in the paper is a division of labour. Sinking and climbing are physically expensive, with oxygen to ration and buoyancy to wrestle, so those themes may be dictated by the body rather than chosen. Down at a stable depth, coasting, a male has room to move. Zeh has suggested that males appear to push their vocal limits at their deepest points to stand out in a crowded breeding ground, according to Oceanographic Magazine.

That is a hypothesis about sexual selection rather than a demonstration of one. Nobody has yet shown a female choosing a mate on the strength of his bottom-phase improvisation.

Sixteen whales, one island chain

Sixteen animals is a respectable haul for tagging work and a modest one for everything else. Zeh says plainly that whether the same pattern holds for humpbacks elsewhere is still open, in coverage by Phys.org.

Good reason for caution there. Humpback song is fashion, and fashion is regional. Michael Noad and colleagues reported in Nature that the song of the east Australian population was replaced wholesale by that of the west coast population, apparently after a small number of foreign singers turned up. Ellen Garland’s group later traced these song revolutions travelling east across the South Pacific, from Tonga to French Polynesia and on to Ecuador, in Royal Society Open Science.

Oxygen and buoyancy work the same way in every ocean. The tunes do not.

A narrow band, badly crowded

The same low-frequency trick that lets humpback song travel so far comes with a ceiling. Coen Elemans and Tecumseh Fitch, voice scientists based in Denmark and Austria, found that baleen whales sing using laryngeal structures found in no other mammal, where air pushed past a large fatty cushion sets it vibrating at very low frequencies. Writing in Nature, they showed those structures also impose hard physiological limits on frequency and depth.

Elemans puts the effective ceiling for communication at about 100 metres, a depth and a frequency range that, as summarised by the University of Southern Denmark, overlap almost exactly with the racket of shipping traffic.

Which gives the Maui result a use beyond curiosity. Underwater microphones are already parked across the world’s oceans. If themes reliably signal depth, a recording alone starts to reveal where in the water column an animal was and what it was doing, a baseline against which disturbance can be measured.

All that singing needs the right slice of water, the right depth, the right band of frequencies. The whales booked that stage first. Nobody told the container ships.