For measuring distances the knowledge of absolute magnitude or luminosity is often crucial, especially for very big distances. Unfortunately we can't measure the diameter of far distant objects and calculate and derive absolute magnitude due to resolution limits.

That's why objects or better named states in the life cycle of specific objects, like Type Ia supernovas, are so important.

What additional objects do we know of sharing this property? Are there objects theoretically predicted to have a absolute magnitude but until now not yet discovered? Name the object and the spectral range of emitted light or particles.


The jargon for what you are looking for is "standard candles": things whose luminosities we can determine without knowing their distance. They are of particular interest to astronomers because they can be used to measure distances.

There are many such objects, but all of them should be treated with some caution. In no case is our knowledge of the luminosity perfect, and in many cases there is large intrinsic scatter. Generally, our knowledge is not of the form "all objects of type x have luminosity y", but more of the form "for objects of type x, the luminosity is correlated with parameters a, b, and c according to complicated equation foo." The physical origin of complicated equation foo is much better understood in some cases than in others, and in all cases needs to be empirically calibrated. Particularly if the physical origin of the correlation is poorly understood, we may not know if or how the calibration changes with the age of the universe. Because we see very distant objects as they were when the universe was younger, this limits our ability to use them as distance measurements to great distances.

In all cases one needs to be careful to take the redshift into account, as the part of an objects rest spectrum which, say, appears blue nearby, may appear red or even IR when the same object is more distant. (See k-correction.) In many cases, a range of wavelengths may be used (at least in the visual or IR), but the calibration may be different for different rest wavelengths. If you observe the all objects through the same filter, you will be observing different objects at different rest wavelengths.

Here are some standard candles:

  • Cepheid variable stars (see 2000ApJS..128..431F) are very bright, and their luminosity is strongly correlated with their luminosity, making them excellent standard candles.

  • RR Lyrae variable stars also follow such a relationship (2003LNP...635...85B), but are fainter.

  • Type Ia supernova are very bright, and their peak luminosity can be estimated from their change in luminosity over time.

  • The tip of the red giant branch in the HR diagram (2000ApJS..128..431F) is one bright feature of the HR diagram that can be used. Blue supergiants have also been proposed as possible standard candles (see 2003LNP...635..123K).

  • The simple surface brightness of a galaxy is useless as a standard candle: the number of stars per square arcsecond rises as the distance squared, while the luminosity of an individual star falls as the distance squared, so the surface brightness is independent of distance. However, even in a galaxy where the stars are distributed according to some smooth function (as in an elliptical galaxy like M87), the surface brightness isn't perfectly smooth, because the stars are of finite brightness: the stars are randomly distributed according to the smooth function, and by chance some places have more stars than others. The roughness of the galaxy can therefore be used to measure the luminosity weighted mean luminosity of the stars in the galaxy, and this can be used as a standard candle of sorts. This is the "surface brightness fluctuation" (SBF) method of distance measurement, introduced in 1988AJ.....96..807T.

  • Large clusters of galaxies usually have a bright giant elliptical galaxy near the center. These are called "Brightest Cluster Galaxies" (BCGs). BCGs have a fairly consistent luminosity; see 1995ApJ...440...28P.

  • Planetary nebulae can have a wide range of luminosities, but there is a well defined upper limit to how bright they can be (see 1989ApJ...339...39J and associated articles). So, if you measure the number of planetary nebulae in a galaxy as a function of luminosity, the "planetary nebula luminosity function" (PNLF), the cutoff at the bright end can be used as a standard candle.

  • The peak of the globular cluster luminosity function (GCLF) seems to be consistent across different galaxies, so the luminosity at which there are the most globular clusters in a given galaxy can be used as a standard candle. The physical reason for this consistency is not well understood. See 2006AJ....132.2333S.

  • For spiral galaxies, there is a relationship between the the rotation curve and luminosity, the "Tully-Fisher" relation (1977A&A....54..661T). See also the Faber-Jackson relation (1976ApJ...204..668F) and Fundamental plane for elliptical galaxies.

  • There may be a relationship between the radius of the broad line region of an active galactic nucleus and its luminosity. See Watson el al. (2011).

  • 1
    $\begingroup$ Good answer, +1. Another i'd add is that there appears to be a limit to how luminous late-B to F-type hypergiants can be, as the highest-mass stars hit an poorly-understood instability that means they do not evolve redwards across the Hertzsprung-Russell diagram to become A/F hypergiants or red supergiants, and the brightest A/F-type stars all have similar luminosity (log(L/Lsun)~5.8). Observationally this is known as the "Humphreys-Davidson limit", an observation that there are no yellow/red stars as luminous as the brightest blue stars. $\endgroup$ – strmqm Aug 7 '11 at 5:50
  • $\begingroup$ Subtle calls for people to upvote your question might discourage them from doing so... Good, thorough answer though. Every candle I know is on there, plus a few I didn't! $\endgroup$ – Warrick Aug 8 '11 at 18:01

A widely used method to estimate distances to clusters and nearby galaxies is based on morphological features in the Hertzsprung-Russell (basically a plot of stellar luminosity versus effective temperature, or the related color) diagram. The locus of stars converting hydrogen to helium in their cores is particularly well-defined on the HR diagram and is called the Main Sequence. The lower end of the Main Sequence can accurately plotted by using nearby stars whose distances are well determined by stellar parallax or other geometric means. The Main Sequence can be extended to higher luminosities by observing young clusters in our galaxy.

Distances to nearby galaxies can be determined by comparing more luminous features in the HR diagram, such as RR Lyrae stars on the 'Horizontal Branch' (locus of highly evolved stars running out of nuclear fuel). A refined method of using Cepheid variables positions on the HR diagram (then called a period-luminosity-color relation) extends the use of the HR diagram up towards supergiant luminosities.

There are a lot of devils in the details, particularly the effects of the 'metallicity' (abundance of elements other than hydrogen and helium) in the stars, which need to be allowed for.


It's been a while for me but the standard candles include


The first thing and the basis of all distance estimations is parallax! Stars in our neighborhood have a certain parallax seen from the diameter of the Earth's orbit. This makes a rather precise measurement of distance. All stars with a precise parallax can be assigned with an absolute luminosity. To go on further, you have to look for "standard candles" within the set of these stars having a reliably measurable parallax.

Another method is the diffraction pattern of stars (coherence length of the star's light). This can be used to determine the diameter of a star. Diameter (area) and color temperature allow to calculate absolute luminosities.


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.