During the Second World War, engineers in the United States Navy discovered a remarkable oceanographic phenomenon. In certain layers of the deep ocean, sound propagates with extraordinarily low attenuation, to the point that an explosion set off in Hawaii can be detected by hydrophones placed in California, more than four thousand kilometres away. They named this layer the SOFAR channel (SOund Fixing And Ranging) and used it for submarine operations. What they did not know was that blue whales had been exploiting this channel for several million years. Their very low frequency songs travel through the SOFAR over distances that defy imagination, and allow oceanic communication on a planetary scale.
Short answer: blue whales produce songs at frequencies below 20 Hz, in a band where underwater acoustic attenuation is at its lowest. These sounds travel through the SOFAR channel, a layer of water lying between 600 and 1,200 m deep where the speed of sound reaches a minimum. This channel acts as a natural waveguide, trapping the signals and allowing them to travel thousands of kilometres.
The attenuation of a sound in water depends mainly on its frequency. At high frequency (10 kHz and above), absorption by water molecules is strong: the sound loses intensity quickly, carrying no more than a few kilometres at best. At low frequency (below 100 Hz), absorption is almost negligible: the sound can in theory travel thousands of kilometres before dying away.
Blue whales vocalise at around 10 to 40 Hz, precisely within the optimal acoustic window. Their songs lie at the lower limit of human hearing (which begins at about 20 Hz). To hear a recorded blue whale song, it often has to be sped up by a factor of ten, which brings the sound into the audible band and reveals fascinating rhythmic modulations. At their natural frequency, they are felt more than they are heard: a dull vibration in the chest.
The counterpart to this immense range is low spatial resolution. At 20 Hz, the wavelength in water is about 75 metres, far too large to pinpoint a target or avoid an obstacle. This is why very low frequencies are not used for echolocation (hunting), but solely for large-scale communication.
The very low attenuation of low frequencies is compounded by a remarkable oceanographic phenomenon. In the ocean, the speed of sound varies with depth: it depends on temperature and on pressure. Near the surface, the water is warm and the pressure low, so sound travels relatively fast (1,530 m/s on average). At depth, temperature falls, but pressure rises. At around 600-1200 metres, these two opposing effects produce a local minimum in the speed of sound.
This layer of minimum speed is precisely the SOFAR channel. Its remarkable property is that it works as a natural waveguide. Any sound wave emitted within this layer is bent by refraction towards the centre of the channel: if it tends to rise, the faster speed in the layers above turns it back downwards; if it tends to sink, the faster speed in the layers below turns it back upwards. The result: the acoustic energy stays confined within the channel, touching neither the surface nor the seabed. It travels horizontally without attenuating as it would in a free medium.
During the Cold War, the United States Navy installed the SOSUS network (Sound Surveillance System), hydrophones placed within the SOFAR along several coastlines, to track Soviet submarines. This network, opened gradually to civilian research since 1991, has allowed scientists to listen to the SOFAR across entire ocean basins, and to quantify the real range of blue whale songs.
Only males sing. The songs are made up of tonal units a few seconds long (generally 10-30 seconds), repeated in sequences that can last for hours, or even days. Each population has its own dialect, strikingly stable across the decades but changing very slowly. Distinct songs are recognised for the blue whales of the Northeast Pacific, the Northwest Pacific, the Southeast Pacific, the Antarctic, the North Atlantic and the Indian Ocean, each with its own acoustic signature.
The exact function of the songs remains debated. The main hypotheses: a reproductive signal directed at females (analogous to the songs of male birds), a marker of population identity, communication between males dispersed across the feeding grounds, very long range detection of conspecifics. Probably several functions at once.
One of the most intriguing observations was made by Mark McDonald, John Hildebrand and their team at the Scripps Institution. Analysing decades of recordings, they showed in a study published in 2009 (Endangered Species Research) that the centre frequency of blue whale songs has fallen significantly and steadily over recent decades, by about 30% in all, in every population studied. This phenomenon, almost without precedent in the animal kingdom, calls for an explanation.
Several hypotheses have been put forward. The first: compensation for anthropogenic noise. Lower frequencies are less masked by shipping noise (which dominates above 50 Hz). But this hypothesis does not account well for the fact that the decline also affects populations relatively isolated from shipping traffic. A second hypothesis: demographic change. The world population of blue whales is gradually recovering after the collapse of industrial whaling, and dominant individuals are on average larger. Larger males naturally produce lower pitched songs. A third hypothesis: reproductive competition: singing lower would signal greater size, attractive to females, and sexual selection would push frequencies ever lower. No hypothesis commands consensus, and the question remains open.
The SOSUS recordings have made it possible to measure the range of the songs directly. Under favourable conditions, blue whale songs are detectable more than 2,500 kilometres from where they are produced. In theory, in a silent pre-industrial ocean, two blue whales could communicate from one continent to another. A whale off California could have heard a whale off Hawaii. A whale in the North Atlantic could have communicated with a whale near the African coast.
Today, shipping noise, concentrated in exactly the same low frequency band, drastically reduces that effective range. In areas of heavy traffic it can be cut by a factor of ten or more. See our page on the impact of ocean noise. This is a silent but profound transformation of the communication landscape of the largest animals that have ever lived.
