Humpback Whale Song

The word 'song' describes the pattern of regular and predictable sounds made by some whales, especially the humpback. Humpback whale song is an extraordinary example of vocal cultural behaviour, and some regard it as the most sophisticated acoustic display in the animal kingdom.

The display is produced by males and is socially learned from conspecifics, evolving slowly under sexual selection. Its architecture is nested: the 'unit' is the shortest sound to the human ear, a few units combine into a 'phrase', repeated phrases make a 'theme', and multiple themes compose a song, usually 5 to 30 minutes long. At any one time most males within a population sing the same version of the song.

Singing generally happens on low-latitude breeding grounds, where whales migrate in winter from higher-latitude feeding grounds. Males also sing on migration and, to a lesser extent, on feeding grounds. Humpback whales show strong maternally directed site fidelity to both breeding and feeding grounds.

The songs of male humpback whales are a rare example among animals of social learning between entire populations. Six distinct song types transmitted from the east Australian to New Caledonian populations between 2009 and 2015 were analysed using fine-scale song features. New Caledonian whales learned each song type with high accuracy, regardless of how complex the pattern was. Learning errors occurred more often in progressively changing 'evolutionary' songs than in rapidly changing 'revolutionary' ones. Fine-scale similarity in song patterns and structural features is what clarifies how accurately songs are learned during inter-population transmission.

Song revolutions spread much faster than the gradual accumulation of small modifications that characterises song evolution. In Okinawa, matched song themes showed minimal evolution between 2011 and 2012, while the 2013 song was more distinct, with singers performing both new and evolved versions of themes.

Vocal learning in this species cannot be studied in the laboratory, so learning is inferred from song complexity and patterns of transmission. Individual-based cultural evolutionary simulations were used to model the entire Southern and Northern Hemisphere humpback populations. The model combined song mutation with patterns of contact among populations and was compared against theme sharing measured in South Pacific populations.

Male humpback whales have been called 'inveterate composers' of songs 'strikingly similar' to human musical traditions. It has been suggested that humpback songs communicate male fitness to females, although that explanation has been challenged on various grounds. The complex sounds of the humpback whale, and of some blue whales, are believed to be used primarily in sexual selection. Song may also be multifunctional, serving several purposes in humpback reproduction.

The song is built from nested elements. A few units combine into a phrase, and repeated phrases make a theme; multiple themes compose a song, usually 5 to 30 minutes long. The complexity of the sequence of units, and small variations within the units themselves, could in theory code information that communicates fitness to both males and females.

Most males within a population sing the same version of the song at any one time. Populations can also add changes and embellishments to songs in the form of themes, which appear to be learned as distinct segments. Song revolutions spread much faster than the gradual accumulation of small modifications that characterises song evolution.

Learning errors occurred more often in progressively changing 'evolutionary' songs than in rapidly changing 'revolutionary' ones. Behaviours can spread by social learning horizontally and vertically, within and between generations, or obliquely between unrelated conspecifics of different generations. Six distinct song types transmitted from the east Australian to New Caledonian populations between 2009 and 2015 were analysed using fine-scale song features.

In Okinawa matched song themes showed minimal evolution between 2011 and 2012, while the 2013 song was more distinct, with singers performing both new and evolved versions of themes. Song complexity scores fell and then rose over the Okinawa study period. Within an ocean basin, song similarity between breeding populations can reveal how connected those populations are. Many themes were shared across the North Pacific, adding to evidence of a single panmictic song lineage across that basin, in contrast to the South Pacific. Compared with the South Pacific, North Pacific populations show a relatively high degree of song sharing across breeding grounds, within and between seasons, indicating a high level of mixing. In the northwestern Pacific there has been very little information on song dynamics and linkages across the basin.

The architecture of the song follows a strict hierarchy: units combine into phrases, phrases into themes, and themes into the song itself. Cultural transmission, understood as the social learning of information or behaviour from conspecifics within a group, is what produces behaviour specific to a group.

During winter humpback whales converge on tropical breeding grounds characterised by shallow, warm seas.

A 2018 acoustic survey by an autonomous Wave Glider from Hawaii eastward toward Mexico, at breeding latitudes and in peak season, found singing whales almost continuously out to mid-ocean, the first evidence of that tropical deep-water distribution. Some humpback populations swim 5,000 miles from tropical breeding grounds to colder, more productive feeding grounds.

The song is a sexually selected display, and it is socially learned from conspecifics rather than inherited as a fixed repertoire. Its architecture is nested: the 'unit' is the shortest sound to the human ear, a few units combine into a 'phrase', repeated phrases make a 'theme', and multiple themes compose a song usually 5 to 30 minutes long. At any one time most males within a population sing the same version. The term 'song' itself describes the pattern of regular and predictable sounds made by some whales, especially the humpback, and males have been called 'inveterate composers' of songs 'strikingly similar' to human musical traditions.

Two modes of change coexist. Revolutions spread much faster than the gradual accumulation of small modifications that characterises evolution. Six distinct song types transmitted from the east Australian to New Caledonian populations between 2009 and 2015 were analysed using fine-scale song features; New Caledonian whales learned each type with high accuracy regardless of how complex the pattern was, and learning errors occurred more often in progressively changing 'evolutionary' songs than in rapidly changing 'revolutionary' ones. Fine-scale similarity in song patterns and structural features is what clarifies how accurately songs are learned during inter-population transmission.

In Okinawa matched song themes showed minimal evolution between 2011 and 2012, while the 2013 song was more distinct, with singers performing both new and evolved versions of themes. Because vocal learning in this species cannot be studied in the laboratory, learning is inferred from song complexity and patterns of transmission. Individual-based cultural evolutionary simulations have modelled the entire Southern and Northern Hemisphere humpback populations, combining song mutation with patterns of contact among populations and comparing the output against theme sharing measured in South Pacific populations.

The songs of male humpback whales are a rare example among animals of social learning between entire populations, and the behaviour is regarded as an extraordinary case of vocal cultural behaviour. What the song is for remains contested: it has been suggested that it communicates male fitness to females, although that explanation has been challenged on various grounds, and the song may be multifunctional, serving several purposes in reproduction. The complex sounds of the humpback, and of some blue whales, are believed to be used primarily in sexual selection.

The complex sounds of the humpback whale, and of some blue whales, are believed to be used primarily in sexual selection.

Under that model, both direct echoes from targets and modulations in song-generated reverberation can inform a singer's decisions about mating, travelling and foraging.

Fifty years of field research have established that humpback song plays a critical role in the species' mating system.

The song itself is built from nested elements: the unit is the shortest sound to the human ear, a few units combine into a phrase, and repeated phrases make a theme, with multiple themes composing a song usually 5 to 30 minutes long. The word 'song' describes this pattern of regular and predictable sounds made by some whales, especially the humpback. Male humpback whales have been called 'inveterate composers' of songs 'strikingly similar' to human musical traditions.

Singing is a socially learned, sexually selected display that evolves slowly, and most males within a population sing the same version of the song at any one time. Singing generally happens on low-latitude breeding grounds, where whales migrate in winter from higher-latitude feeding grounds; males also sing on migration and, to a lesser extent, on feeding grounds. Humpback whales show strong maternally directed site fidelity to both breeding and feeding grounds.

Song revolutions spread much faster than the gradual accumulation of small modifications that characterises song evolution. The songs of male humpback whales are a rare example among animals of social learning between entire populations, and humpback whale song is an extraordinary example of vocal cultural behaviour.

Six distinct song types transmitted from the east Australian to New Caledonian populations between 2009 and 2015 were analysed using fine-scale song features. New Caledonian whales learned each song type with high accuracy, regardless of how complex the pattern was, and learning errors occurred more often in progressively changing 'evolutionary' songs than in rapidly changing 'revolutionary' ones. Fine-scale similarity in song patterns and structural features is what clarifies how accurately songs are learned during inter-population transmission. In Okinawa, matched song themes showed minimal evolution between 2011 and 2012, while the 2013 song was more distinct, with singers performing both new and evolved versions of themes.

Because vocal learning in this species cannot be studied in the laboratory, learning is inferred from song complexity and patterns of transmission. Individual-based cultural evolutionary simulations were used to model the entire Southern and Northern Hemisphere humpback populations, combining song mutation with patterns of contact among populations and comparing the results against theme sharing measured in South Pacific populations.

Such behaviours can propagate horizontally and vertically, within and between generations, or obliquely between unrelated conspecifics of different generations.

Vocal learning in this species cannot be studied in the laboratory, so learning is inferred from song complexity and patterns of transmission.

The songs of male humpback whales are a rare instance among animals of social learning that spans entire populations. Song is likewise regarded as an extraordinary case of vocal cultural behaviour. Because vocal learning in this species cannot be studied in the laboratory, learning is inferred from song complexity and from patterns of transmission. Fine-scale similarity in song patterns and structural features is what clarifies how accurately songs are learned when they pass between populations, and learning accuracy is high regardless of song complexity. Learning errors occurred more often in progressively changing evolutionary songs than in rapidly changing revolutionary ones.

Documented cultural revolutions have been recorded in South Pacific and Southern Hemisphere populations. Individual-based cultural evolutionary simulations were used to model the entire Southern and Northern Hemisphere humpback populations.

Birdsong remains the most studied example of vocal learning, with every experimentally tested songbird shown to learn socially the song components used in resource defence and breeding. For songbirds, evidence of ubiquitous vocal learning comes instead from field studies inferring learning from geographic variation in song or from changes in song-type frequency over time.

In northern populations the complex songs show long-lasting traditions that slowly evolve. The South Pacific follows a different logic: periodic revolutions occur when songs are adopted from neighbouring populations and spread rapidly.

Compared with the South Pacific, North Pacific populations show a relatively high degree of song sharing across breeding grounds, within and between seasons, indicating a high level of mixing. Many themes were shared across the North Pacific, adding to evidence of a single panmictic song lineage across that basin, in contrast to the South Pacific.

The revolutions documented in the South Pacific belong to the Southern Hemisphere populations, where songs are taken up from neighbouring populations and spread rapidly. Understanding how such transmission works has required indirect methods: vocal learning in this species cannot be studied in the laboratory, so learning is inferred from song complexity and patterns of transmission. Fine-scale similarity in song patterns and structural features is what clarifies how accurately songs are learned during inter-population transmission, and the evidence indicates that populations do not need to reduce complexity in order to learn song patterns accurately.

Much of this picture rests on simulation. Individual-based cultural evolutionary simulations were used to model the entire Southern and Northern Hemisphere humpback populations. The model combined song mutation with patterns of contact among populations and was compared against theme sharing measured in South Pacific populations.

Coverage is uneven across the northern basin. In the northwestern Pacific there has been very little information on song dynamics and linkages across the basin. In the eastern North Pacific two breeding grounds, along Mexico's shoreline shallows and in mid-ocean around Hawaii, are separated by 4,500 to 6,000 km of deep ocean basin.

Six distinct song types transmitted from the east Australian to the New Caledonian populations between 2009 and 2015 were analysed using fine-scale song features. The New Caledonian whales learned each song type with high accuracy regardless of its complexity. Learning errors occurred more often in progressively changing 'evolutionary' songs than in rapidly changing 'revolutionary' ones.

Fine-scale similarity in song patterns and structural features is what clarifies how accurately songs are learned during inter-population transmission. Populations do not need to reduce complexity in order to learn accurately. They can add changes and embellishments as themes learned as distinct segments.

A separate line of evidence comes from Okinawa, where matched song themes showed minimal evolution between 2011 and 2012 while the 2013 song was more distinct, with singers performing both new and evolved versions of themes. Over that study period song complexity scores fell and then rose.

Matched song themes showed minimal evolution between 2011 and 2012, indicating a high degree of continuity in the song over that interval. The 2013 song was more distinct, and singers were observed performing both new and evolved versions of themes. Song complexity scores fell and then rose over the Okinawa study period.

In the northwestern Pacific there has been very little information on song dynamics and linkages across the basin.

The Okinawa record sits within a wider picture of how humpback song is built and transmitted. The lowest level of the song is the unit, the shortest sound to the human ear. A few units combine into a phrase, and repeated phrases make a theme; multiple themes compose a song, usually 5 to 30 minutes long. Male humpback whales sing a slowly evolving, sexually selected song display that is socially learned from conspecifics, and most males within a population sing the same version of the song at any one time.

Singing generally happens on low-latitude breeding grounds, where whales migrate in winter from higher-latitude feeding grounds. Males also sing on migration and, to a lesser extent, on feeding grounds. Humpback whales show strong maternally directed site fidelity to both breeding and feeding grounds.

Song change takes more than one form. Song revolutions spread much faster than the gradual accumulation of small modifications that characterises song evolution. The songs of male humpback whales are a rare example among animals of social learning between entire populations, and the song is an extraordinary example of vocal cultural behaviour.

Transmission between populations has been measured directly. Six distinct song types transmitted from the east Australian to New Caledonian populations between 2009 and 2015 were analysed using fine-scale song features. New Caledonian whales learned each song type with high accuracy, regardless of how complex the pattern was. Learning errors occurred more often in progressively changing 'evolutionary' songs than in rapidly changing 'revolutionary' ones, and fine-scale similarity in song patterns and structural features is what clarifies how accurately songs are learned during inter-population transmission.

Because vocal learning in this species cannot be studied in the laboratory, learning is inferred from song complexity and patterns of transmission. Individual-based cultural evolutionary simulations were used to model the entire Southern and Northern Hemisphere humpback populations; the model combined song mutation with patterns of contact among populations and was compared against theme sharing measured in South Pacific populations.

The word 'song' describes the pattern of regular and predictable sounds made by some whales, especially the humpback. Male humpback whales have been called 'inveterate composers' of songs 'strikingly similar' to human musical traditions. It has been suggested that humpback songs communicate male fitness to females, although that explanation has been challenged on various grounds, and song may also be multifunctional, serving several purposes in humpback reproduction.

Individual-based cultural evolutionary simulations have been used to model the entire Southern and Northern Hemisphere humpback populations.

Within this framework the model combined song mutation with patterns of contact among populations.

The model's output was compared against theme sharing measured in South Pacific populations.

Observational comparison across ocean basins sharpens the picture: compared with the South Pacific, North Pacific populations show a relatively high degree of song sharing across breeding grounds, within and between seasons, indicating a high level of mixing.

The song itself is built from nested elements: the lowest level is the 'unit', the shortest sound to the human ear. A few units combine into a 'phrase', and repeated phrases make a 'theme'; multiple themes compose a song, usually 5 to 30 minutes long. The display is a slowly evolving, sexually selected behaviour that males learn socially from conspecifics, and most males within a population sing the same version of the song at any one time.

Singing is concentrated on low-latitude breeding grounds, where whales migrate in winter from higher-latitude feeding grounds. Males also sing on migration and, to a lesser extent, on feeding grounds. Against this background of movement, humpback whales show strong maternally directed site fidelity to both breeding and feeding grounds.

Two modes of change coexist. Song revolutions spread much faster than the gradual accumulation of small modifications that characterises song evolution. Because vocal learning in this species cannot be studied in the laboratory, learning is inferred from song complexity and patterns of transmission.

Fine-scale analysis supplies the measurements. Six distinct song types transmitted from the east Australian to New Caledonian populations between 2009 and 2015 were analysed using fine-scale song features. New Caledonian whales learned each song type with high accuracy, regardless of how complex the pattern was, and learning errors occurred more often in progressively changing 'evolutionary' songs than in rapidly changing 'revolutionary' ones. Fine-scale similarity in song patterns and structural features is what clarifies how accurately songs are learned during inter-population transmission. At a single locality, matched song themes in Okinawa showed minimal evolution between 2011 and 2012, while the 2013 song was more distinct, with singers performing both new and evolved versions of themes.

The songs of male humpback whales are a rare example among animals of social learning between entire populations, and the behaviour has been called an extraordinary example of vocal cultural behaviour. Recording and characterising the flow of socially learned behaviours makes it possible to census the level of connectivity among populations. The word 'song' describes the pattern of regular and predictable sounds made by some whales, especially the humpback, and males have been called 'inveterate composers' of songs 'strikingly similar' to human musical traditions.

It has been suggested that humpback songs communicate male fitness to females, although that explanation has been challenged on various grounds, and song may also be multifunctional, serving several purposes in humpback reproduction.

Ambient sound levels in humpback habitat shift with the chorus itself: noise from singing decreased during daytime nearshore while rising offshore. Such changes in ambient sound levels, produced by many males singing concurrently but asynchronously, have been linked to changes in overall local whale abundance in Hawaii.

In the eastern North Pacific two breeding grounds lie along Mexico's shoreline shallows and in mid-ocean around Hawaii, separated by 4,500 to 6,000 km of deep ocean basin. A 2018 acoustic survey by an autonomous Wave Glider tracked from Hawaii eastward toward Mexico at breeding latitudes and in peak season. It found singing whales almost continuously out to mid-ocean, the first evidence of that tropical deep-water distribution.

The song itself is built from nested levels. Its lowest element is the unit, the shortest sound to the human ear. A few units combine into a phrase, repeated phrases make a theme, and multiple themes compose a song, usually 5 to 30 minutes long. Most males within a population sing the same version of the song at any one time.

Singing is a sexually selected display that males learn socially from conspecifics, and it evolves slowly. It generally happens on low-latitude breeding grounds, where whales migrate in winter from higher-latitude feeding grounds. Males also sing on migration and, to a lesser extent, on feeding grounds. The display may be multifunctional, serving several purposes in humpback reproduction, and some regard it as the most sophisticated acoustic display in the animal kingdom. The word 'song' describes the pattern of regular and predictable sounds made by some whales, especially the humpback, and males have been called 'inveterate composers' of songs 'strikingly similar' to human musical traditions.

Song is an extraordinary case of vocal cultural behaviour and a rare animal example of social learning between entire populations. Revolutions spread much faster than the gradual accumulation of small modifications that characterises evolution. Because vocal learning in this species cannot be studied in the laboratory, learning is inferred from song complexity and patterns of transmission.

Six distinct song types transmitted from the east Australian to New Caledonian populations between 2009 and 2015 were analysed using fine-scale song features. New Caledonian whales learned each song type with high accuracy, regardless of how complex the pattern was, while learning errors occurred more often in progressively changing 'evolutionary' songs than in rapidly changing 'revolutionary' ones. Fine-scale similarity in song patterns and structural features is what clarifies how accurately songs are learned during inter-population transmission. In Okinawa, matched song themes showed minimal evolution between 2011 and 2012, while the 2013 song was more distinct, with singers performing both new and evolved versions of themes. Humpback whales also show strong maternally directed site fidelity to both breeding and feeding grounds.

Individual-based cultural evolutionary simulations were used to model the entire Southern and Northern Hemisphere humpback populations. The model combined song mutation with patterns of contact among populations and was compared against theme sharing measured in South Pacific populations.