mariuscomper.ukRO

North Pacific · recordings 1963–2019

The 52 Hz whale, ‘the loneliest in the world’, and the blue whales singing ever deeper since 1963

From 1989, US Navy hydrophones in the North Pacific picked up a whale song unlike any other, at 52 hertz. The press later called it ‘the loneliest whale in the world’. Meanwhile, something else was found: from 1963 to 2019 the song of the blue whales of the northeast Pacific dropped in pitch by a third, and no explanation proposed so far fits all the measurements.

How low were blue whales singing in the year you were born?

Type your year of birth. The page takes a real NOAA recording from the northeast Pacific and tunes it to the note measured that year, then to the 2019 note. Blue-whale song is so deep the ear barely catches it, so everything here plays 20 times higher than in the sea and, like tape run at another speed, 17 to 27 times faster depending on the year: a 20-second call lasts about a second. Headphones help. The spectrogram below the buttons draws the sound as it plays, on the sea’s hertz scale.

From 1963 to 2019 the base note fell from about 21.9 Hz to 14.2 Hz: 7.5 semitones, between a fifth and a minor sixth. Type your year of birth to see where you sit on the line.

Base note of the B call of northeast Pacific blue whales, 1963–2019: from about 21.9 Hz to 14.2 Hz14161820221960197019801990200020102020Hz1963: 21.90 Hz1971: 20.47 Hz1974: 20.30 Hz1980: 19.20 Hz1985: 18.50 Hz1987: 17.80 Hz1990: 17.70 Hz1993: 17.50 Hz1993: 17.40 Hz1993: 17.90 Hz1994: 17.50 Hz1996: 17.07 Hz1997: 16.83 Hz1997: 16.90 Hz1998: 16.83 Hz2000: 16.23 Hz2001: 16.00 Hz2002: 15.97 Hz2003: 15.67 Hz2005: 15.53 Hz2006: 15.33 Hz2007: 15.33 Hz2008: 15.17 Hz2009: 15.10 Hz2010: 14.97 Hz2011: 14.87 Hz2012: 14.73 Hz2013: 14.67 Hz2014: 14.53 Hz2015: 14.47 Hz2016: 14.43 Hz2017: 14.30 Hz2018: 14.30 Hz2019: 14.20 Hz1963 · 21.9 Hz2019 · 14.2 Hz
Dots are published measurements, one per year (McDonald and colleagues, 2009, up to 2008; Rice and colleagues, 2022, from 2009). The line follows the trend fitted to all 23 of McDonald’s measurements (1963–2008), drawn up to 2005, then Rice’s yearly means. Values are the base note of the B call, a third of the harmonic the researchers measured.

What is the 52 Hz whale?

During the Cold War, the US Navy laid arrays of hydrophones on the Pacific floor, called SOSUS, to listen for submarines. Biologists later got to work on some of the recordings. William Watkins and his team at Woods Hole, Massachusetts, found in them a series of calls they could not match to any species: tones of three to ten seconds, in groups, strongest at around 52 hertz. They first noticed them in 1989; from 1992 they tracked them season by season.

12 seasons tracked, 1992–2004

11,062 km longest track, in a single season

47 km a day, on average

August–February when it was heard; most often December and January

In 2004 the team published what 12 years had shown. Only one series of calls was ever heard at a time, never two overlapping, which suggested a single whale. The tracks, from 708 to 11,062 kilometres a season, did not follow the blue, humpback or fin whales heard on the same hydrophones. And one short sentence that still stands: the species producing these calls is unknown. Watkins died in September 2004, before the paper came out.

Both recordings are NOAA PMEL's, from the northeast Pacific, played here 20 times higher than in the sea. The 52 Hz clip is undated; its note measures about 50 Hz.

Is it really the loneliest whale in the world?

The word ‘lonely’ is not in the 2004 paper. The story is: in December 2004 The New York Times wrote about an animal roaming the ocean year after year and getting no reply. The phrase ‘the loneliest whale in the world’ spread later; the oldest dated use we found is from 2012, and its exact origin is unknown.

What the data support is less dramatic. Its voice was the only one of its kind on the Navy arrays, but nothing shows the whale is cut off: 52 hertz is not a frequency other whales cannot hear. Christopher Clark of Cornell University, who recorded it in 1993, said in 2015 that blue, fin and humpback whales ‘can hear this guy, they’re not deaf’; the voice is just odd. Kate Stafford, who studies blue whales, noted that it had survived at least 12 years in good shape, and that its seasons and migration pattern match those of blue whales, even though Watkins found its individual tracks unrelated to theirs.

The rest is hypothesis. That it is a hybrid of a blue and a fin whale, as often written, has not been confirmed: as of 2015, nobody had managed to record a hybrid’s call. That several whales may share the voice was suggested by John Hildebrand of Scripps in a 2015 interview: calls of the same type, found in 2010 recordings off California, reportedly appeared at the same time on two widely separated sensors. The observation has not been published in a scientific journal.

Why are blue whale songs getting deeper?

The 52 Hz story hid the other one. In 2009 Mark McDonald, John Hildebrand and Sarah Mesnick gathered blue-whale recordings from around the world, some from the 1960s, and measured the note of each song. In the northeast Pacific the base note of the B call fell from 21.9 Hz in 1963 to 15.2 Hz in 2008, by 31%. All seven song types with measurements across several years were falling, each at its own rate. And within one year, different whales in the same region sang the same note, to within 3%.

  1. Northeast Pacific1963–2008 · 21.9 → 15.2 Hz−6.8%
  2. Northwest Pacific1968–2001 · 25 → 19.45 Hz−6.7%
  3. Southern Ocean1995–2005 · 28.5 → 26.9 Hz−5.6%
  4. North Atlantic1959–2004 · 23 → 17.6 Hz−5.2%
  5. Northern Indian Ocean1984–2002 · 116 → 106 Hz−4.8%
  6. Southwest Pacific1964–1998 · 30.8 → 25.8 Hz−4.8%
  7. Southeast Indian Ocean1993–2000 · 19.5 → 19 Hz−3.7%
How far each song type’s note fell per decade, between its first and last measured year. Source: McDonald and colleagues, 2009, Table 1. Short spans and thin samples (southeast and north Indian Ocean, Southern Ocean) are the least certain.

The California measurements went on. Rice and colleagues published 2006–2019 in 2022: the note keeps falling, more slowly, about 0.27 Hz a year at the measured harmonic against 0.4 up to 2008. In 2019 the base note was 14.2 Hz, below the 20 Hz where human hearing usually begins.

The explanations tested so far do not all fit:

Ship noise

Background noise in the northeast Pacific rose by about 12 decibels in four decades, right in the song’s band. But to escape it, whales should have gone up, not down. Not the main cause, it seems (McDonald, 2009).

Bigger whales after whaling

A bigger whale sings lower. But blue whales reach almost full weight by about eight, so body size should have returned to normal long ago, while the note has kept falling for more than half a century. A secondary factor at most.

A warming ocean

By McDonald’s estimate, four decades of warming changed the speed of sound in the upper 700 m of the ocean by under 0.02%, over a thousand times less than the drop in the note, and it is not clear how it could shift the note a whale produces.

More whales

The hypothesis McDonald called ‘intriguing and testable’: after whaling, the survivors were few and far apart, and males had to sing louder, so higher, to be heard. With more whales around, they can afford a lower note. Rice (2022) points out what it misses: other studies found the note also falls within each season, and the link between loudness and pitch has not been seen in blue whales.

Rice and colleagues raise a stranger idea too: the whales may not even perceive the small, steady adjustments they make through each season, and the descent may be an unintended side effect of the way they keep their songs in tune with each other. Their conclusion: no theory so far explains all of it.

How many blue whales were killed?

The oldest blue-whale song recordings in McDonald’s study date from 1959 (North Atlantic), and those of the northeast Pacific song from 1963, just as industrial whaling was ending: blue whales were protected from 1966, though Soviet ships kept killing them illegally into the early 1970s. In the 20th century 379,185 blue whales were caught, by the count of Rocha and colleagues (2014), almost all in the southern hemisphere.

One dot is 100 blue whales caught between 1900 and 1999. The 88 lit ones are from the North Pacific (8,838), which includes the waters of the song on this page; 67 are from the North Atlantic (6,699); the rest, 363,648, from the southern hemisphere.

How many blue whales live in the northeast Pacific today is not known for sure. The latest photo-identification estimate for the US West Coast is 1,898 in 2018, and NOAA’s 2023 report says the population trend is unknown and that ships probably kill more whales than the limit NOAA considers sustainable for this population.

Where is the 52 Hz whale now?

Watkins’s published tracking series ends with the winter of 2003–2004; the 2010 detection was only described in an interview. Whether it is alive is unknown, and so is its sex.

One thing changed in this voice too: over 15 years its note slid from about 52 Hz to close to 50 Hz, according to Watkins. In 2015 Hildebrand expected it to be nearer 47 Hz by then, though no recording had been identified for a few years. In proportion, that is less than the blue whales dropped over the same years: about 4% against about 12%. Whether the cause is the same, nobody has studied.

We do not know whether anyone ever answered it. We do know it did not sing into a silent ocean: every year, the blue whales of a region tune to one another, and their note fell over more than five decades of measurements. The voice called the loneliest in the world went down too.

Method and sources

Your year’s note. For 1963, the earliest measured year, we use the measurement (24 calls from three whales). For 1964–2005 we read the value off the least-squares line through all 23 measurements in Table S1 of McDonald and colleagues (2009): some years rest on a single call, so the line is steadier than any single point. For 2006–2019 we use the yearly means from Rice and colleagues (2022), 30 calls each from September–October. The two studies measure at slightly different points of the call (the middle, and 10 seconds in) and agree within 0.1 Hz in the shared years 2006–2008. The researchers measured the third harmonic; we show a third of it, the base note, because that is what the recording sounds on. From 1963 to 2019 the note fell by 35% (7.5 semitones).

The sound. The blue-whale recording is NOAA PMEL’s, from the northeast Pacific, published sped up 10 times. We measured the B call’s base note at the middle of each of the clip’s five calls: about 16.2 Hz in the sea. For each year the page changes the playback speed so the note plays 20 times higher than the note measured that year. Tempo changes with pitch. The gap between two years is exact; the absolute pitch depends on our measurement of the clip, within a few per cent. Clips are loudness-matched (−20 LUFS).

The 52 Hz whale. Facts come from Watkins and colleagues (2004). Quotes from Clark, Stafford and Hildebrand are from interviews published by the BBC in 2015 and The New York Times in 2004; the 2010 detection and the 47 Hz value have not been published in journals. The 1989–2004 percentage comparison uses Watkins’s approximate figures (52 Hz, then close to 50 Hz) and blue-whale values read off the same trend line.

The oceans. Per-decade percentages come from the first and last year of each song type in Table 1 of McDonald (2009); where the table gives two notes, we used the first.

The note for every year, 1963–2019
YearBase note (Hz)From
196321.90measured
196421.67trend line
196521.53trend line
196621.38trend line
196721.23trend line
196821.08trend line
196920.93trend line
197020.78trend line
197120.64trend line
197220.49trend line
197320.34trend line
197420.19trend line
197520.04trend line
197619.90trend line
197719.75trend line
197819.60trend line
197919.45trend line
198019.30trend line
198119.15trend line
198219.01trend line
198318.86trend line
198418.71trend line
198518.56trend line
198618.41trend line
198718.26trend line
198818.12trend line
198917.97trend line
199017.82trend line
199117.67trend line
199217.52trend line
199317.38trend line
199417.23trend line
199517.08trend line
199616.93trend line
199716.78trend line
199816.63trend line
199916.49trend line
200016.34trend line
200116.19trend line
200216.04trend line
200315.89trend line
200415.74trend line
200515.60trend line
200615.37measured
200715.33measured
200815.17measured
200915.10measured
201014.97measured
201114.87measured
201214.73measured
201314.67measured
201414.53measured
201514.47measured
201614.43measured
201714.30measured
201814.30measured
201914.20measured

Sources

More on the site

Published on 2 October 2026.