Singing Stars

Sound, noise, and music analysis can be a confusing thing. It is true that there are now Smartphone apps that can provide a wide range of data concerning the sound level expressed in dB, dBA, and even Leq. There are even apps that can provide frequency analysis replete with Fourier or FFT spectral capabilities.

One of the elements of all sound is its definition- a vibrating source and a medium in which to travel.  (The requirement of a receiver is simplistic and should be removed from the 8th grade science curriculum).  Now we have stars that are emitting “music”.

This recent finding from Dr Andrea Miglio at the University of Birmingham’s School of Physics and Astronomy – that stars sing – may take some explanation and perhaps some playfulness with the definitions of basic science.

Even if a star could emit sound – singing or otherwise – there is no medium in galactic space that could transmit the sound.  Dr. Miglio’s finding is a metaphor for an equivalent peculiar light wave emanation. But let’s take a step back.

It is true that music, noise, and indeed all sound does require a medium in which to travel. This is true except for Star Trek and Battle Star Galactica where the space cruisers and Starships whooshed past us listeners, giving us the sensation of speed and motion. But other than that space is almost a vacuum with very few molecules that can oscillate.

The physical definition of a decibel or dB is 10log(level #1/level #2).  Depending on what reference we choose for (sound) level #2, the decibel can be dBA, dB SPL, or dB HL or, more generally, a decibel is 10log (ratio of two things).  I am 5’9” (on a good day when I stand up correctly) and if the reference I choose is also 5’9” then I stand 0 dB tall.  A decibel can be thought of as any relationship between two things and the reference relationship can be arbitrary. Many physical measures that are used are similar in this respect. For example, if I say that it is 30 degrees out today, should I wear a jacket and gloves or a bathing suit?  Is this 30 degrees C or 30 degrees F? And to really confuse things, even frequency differences can be expressed in decibels (10log(f1/f2).  Scientists refer to this as critical ratios and can be expressed in dB or in Hz.

Well, so much for the (sound) level of a vibrating star.

What about the frequency? Frequency is unlike the decibel in the sense that no reference value needs to be specified. It is merely the number of complete wavelength cycles passing a point in space at any one second interval. A frequency of 1000 Hz means that there are 1000 complete vibrations of a molecule every second. Because no reference is used for this physical measure, a frequency is a frequency is a frequency.

But back to stars. Dr. Miglio and other astronomers have known for ages that stars are mostly gas filled – at least near the surface and they are probably liquid or perhaps even solid at the core. Since gas is compressible, stars can vibrate with a subsequent changing of their apparent brightness.  This fluctuation can be thought of as a frequency of oscillation (of its brightness) and can be translated into an “equivalent” frequency fluctuation of the sound, and this is what her research group has found.  It is not really a “finding”, but more of a re-interpretation of observed data.

Based on their gaseous content, mass, and gravitational characteristic, the way that certain stars oscillate gives rise to the singing melody of the stars.

And good luck trying to figure out the sound level of the singing – without a stellar reference, your guess is as good as mine!

About Marshall Chasin

Marshall Chasin, AuD, is a clinical and research audiologist who has a special interest in the prevention of hearing loss for musicians, as well as the treatment of those who have hearing loss. I have other special interests such as clarinet and karate, but those may come out in the blog over time.


  1. Hi Robert: Thank you for your comments. I, like many of my audiology colleagues, get the 3 dB/6 dB issue interchanged. And I definitely have scrubbed the word “intensity” from my vocabulary. I don’t work in the realm of noise control, so don’t really need to worry about vector quantities and intensity.

    The idea of using a dB value to express something other than sound level was first brought to my attention in grad school. We were talking about critical bandwidth- a psycho-physical concept and instead of speaking in terms of Hz, the professor started to talk about critical ratios in terms of dB. The two are equivalent, and that is where I first heard a frequency range expressed in dB. I put up my hand to inquire about this, and I was told that “my incompetence was only exceeded by my lack of originality!”… you have to love the Canadian school system!

  2. A correction which does not contradict the message of your post, but here it is for the record.
    Definition of a decibel or dB is 10log(level #1/level #2) if level#1 and level#2 have units of power, such as watts. In the case of electrical power, P = V*V/R, where P is power in watts, V is potential in Volts and R is resistance in Ohms.
    In decibels, dB = 10log(P1/P2) = 10log((V1*V1)/R)/(V2*V2/R)). Cancel the Rs from numerator and denominator and use some algebra on the “squaring” operation to get
    dB = 10log(V1/V2)^2. Finally use the property of logarithms and exponents to get, for voltage ratios: dB = 20log(V1/V2). Voltage gain of 2 volts per volt, in dB is 20log(2/1) = slightly more than 6 dB.
    Sound power in dB is indeed 10log(level #1/level #2) can be carried over to any level which can be considered power. I never heard frequency ratios expressed in dB but I think it ought to be 20log(freq1/freq2). Likewise for height. I’ve shrunken to 5’6. Compared to your height I would say I am 20log(5.5/5.75) = -3.9 db_Chasin, not -0.19 dB_Chasin.

Comments are closed.