I was reading the recent LIGO paper and one passage stuck out to me:

The system reached a peak gravitational-wave luminosity of $3.6^{+0.5}_{−0.4}× 10^{56}\:\mathrm{erg/s}$, equivalent to $200^{+30}_{−20}M_⊙c^2/\mathrm s$.

Note, in particular, that the radiated power is quoted in $\mathrm{erg/s}$, which seems pretty weird to me, since it is not an SI unit, and it seems to offer very little advantage over the use of joules. Is this a common usage in astrophysics? If so, is there some specific reason for this? (For example, papers using CGS electromagnetic units naturally use ergs for energy measurements, though the use of CGS for anything that actually requires you to write down numbers with units seems to be on a steep decline. This is not the case here, though.) Or is this just some bit of culture that's hard to live down nowadays?

  • 1
    $\begingroup$ I heard of an astrophysicist who likes to say that the SI system is the worst invention ever and the cgs system the best. Since this is objectively false, I can only guess it's a weird culture thing... $\endgroup$
    – Martin
    Feb 12, 2016 at 17:58
  • 1
    $\begingroup$ Related and/or dupe: physics.stackexchange.com/questions/107552/… $\endgroup$
    – Kyle Kanos
    Feb 12, 2016 at 19:31

2 Answers 2


Yes, you commonly see luminosity quoted with two numbers: an occasional frequency range in keV/MeV/GeV and then a luminosity in erg/s. The use of two different energy units in one sentence is its own sort of eyebrow-raiser; Here is a bevy of recent examples, to prove that it's neither a sign of age nor a particularity of the LIGO group: [1], [2], [3], [4], [5], [6], [7], [8].

This course from nrao.edu states that "Most astrophysical theory is done in cgs units, but radio observations are usually reported in mks units since engineers use mks." (Here "mks" is an abbreviated SI system.) Maoz's Astrophysics in a Nutshell also seems to use cgs quoting erg/s, and this textbook claims that it just happens to be the most common in astrophysics.

What's really interesting is that I cannot find any astrophysical journals which require it, but two astrophysical standards organizations forbid it: IAU and AAS. So CGS is persisting in common practice despite institutional pressure rather than because of it.

CGS has persisted in astrophysics and continuum mechanics. The exact reasons are not 100% clear: however in both cases it is nice that it is within a few factors of 10 from the equivalent SI units, which allows you to quickly communicate to the engineers on the team what a wattage is if you have a power output in erg/s, so probably that is a huge factor. This reminds me of the eV unit where it is a compromise between the theorist's requirement that we never use Coulombs and the engineer's requirement that we use volts.

In the latter case of continuum mechanics, it's probably just because cgs gave nice names to viscosity properties, "poise" and "stokes". In the astrophysical case I am less certain, but it seems like it's probably just because occasionally theorists need the Maxwell equations and yet nobody really wants to be bothered by the permeability and permittivity of free space, these being highly unnecessary concepts created by the ubiquity of volts and amps.

  • 2
    $\begingroup$ If by "forbid" you mean the same thing the USA meant when it "forbade" vehicles from exceeding 55 mph from 1973 to 1995 (wink, wink), then yes, the IAS and AAS have "forbidden" the use of CGS units. However, a quick glance at various AAS journals says this is a "wink, wink" kind of prohibition. The IAS merely says that use of CGS units is deprecated rather than forbidden. $\endgroup$ Feb 12, 2016 at 18:53
  • $\begingroup$ Historical conventions are hard to break... $\endgroup$
    – Kyle Kanos
    Feb 12, 2016 at 19:30
  • $\begingroup$ I suspect that Schrawzchild's book may have something to do with it. It dates from a period when cgs units were very common throughout science and remained a go-to text for decades into the (nominal) SI era. $\endgroup$ Feb 12, 2016 at 20:43
  • $\begingroup$ E&M definitely plays a huge role in astrophysics' reasoning: the vast majority of matter in the universe is ionized plasma, generating and responding to electromagnetic fields. $\endgroup$
    – user10851
    Feb 23, 2016 at 2:22
  • $\begingroup$ The thing is, that once you're brain works in CGS, it is hard to switch. When I see values quoted in SI, I fail to intuitively grasp them right away, just because my brain is now wired to CGS. $\endgroup$ Oct 8, 2021 at 4:26

As far as I know from my “Intro to Astronomy” lecture it is a unit they have always used. Same reason they use Parsec, light-year and other things like the OBAFGKMT ordering scheme for spectral classes.

High energy physics has settled on the natural units with $\hbar = c = 1$ and measure all the energies in electron-volt. Most other experimental physics uses the SI system as the definitions of most units make it sensible to actually measure. A unit system based on the gravitational constant, which is not known that precisely does not make that much sense from an engineering standpoint.

Astronomy is a field with a long heritage, therefore they stick to their units, I believe. So I would assume that most literature and papers (and software) is written using those units, the work of changing that in the whole field will probably exceed the benefits of using a unified unit system.

As a theoretician/computational physicist, I happen to work with a lot of natural units or dimensionless numbers anyway, so I do not care that much about particular unit systems any more.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.