Just a few days ago there was a story about a new type of “standard candle” for use in measuring very large astronomical distances. Such standard candles have been lacking when the distances to be measured are a bit more than halfway back to the big bang – around 7 or 8 billion light-years. Consequently, it’s very difficult to obtain reasonably precise data about many things that happened in the early universe, such as the rate at which the universe was expanding then. This is a major problem since phenomena such as dark energy are difficult to theorize about without good data.
Now another type of standard candle has been identified, and it also makes it possible to gauge large distances – even to objects whose light was emitted about 1.5 billion years after the big bang, at a distance of 12 billion light-years (redshift z~4).
In theory, distance should be simple to work out. If you know the intrinsic brightness of an object, a simple measure of its apparent brightness will tell you how far away it is (since brightness falls as an inverse square of its distance).
So in astronomy, the problem of distance is intimately linked to the problem of knowing an object’s intrinsic brightness.
But that’s hard. There’s simply no way to tell the intrinsic brightness of most stars and galaxies and so no way to work out their distance.
In order for a particular type of object to serve as a good standard candle astrophysicists must be able to determine how much energy it actually radiates – how luminous it is at various wavelengths. One way is to have a good supply of such objects readily available and amenable to reasonably precise measurements of the actual luminosity. This was possible, for example, with the original standard candles: Cepheid variable stars, in which there is a predictable relationship between luminosity and period of the variability.
This approach was used with the other recently established type of standard candles: gamma-ray bursts. In these objects it was found (in cases which are relatively nearby) that there’s a good relationship between the total energy output and the pattern of variation over time of the energy flux. That’s similar to the case with Cepheid variables.
The alternative is to have a sufficiently good theoretical understanding of the energy output of the object in order to calculate luminosity from other observable characteristics. This approach is used with Type Ia supernovae. Since we know how their energy is generated, it’s straightforward to calculate their luminosity.
Now there is yet another type of object that may be useful as a standard candle – the nuclei of active galaxies. Observation of nearby examples of such objects resulted in a fairly reasonable understanding of their structure. The central object of an active galactic nucleus is a supermassive black hole. This causes gas and dust in the vicinity to collapse in towards the center, where it’s heated to high temperatures (due to conversion of potential energy into kinetic energy) and forms a very luminous accretion disk. Surrounding this accretion disk at a larger distance from the black hole are thick clouds of colder (but still pretty warm) gas and dust in the form of a torus. This is called the “broad line region” (BLR), because its spectral lines are broadened due to the Doppler shift resulting from its rotation.
The accretion disk is not always directly visible, if it happens to be obscured by the gas and dust of the BLR. However, the BLR is heated and ionized to a plasma by the much hotter accretion disk. Thus the BLR itself is luminous at a range of wavelengths. The luminosity is proportional to the size of the BLR, hence to its radius. But the radius can be estimated, since the BLR is heated by the accretion disk. Fluctuations in radiation from the accretion disk lead to flucuations in the BLR’s radiation, but with a delay due to the finite speed of light – typically from 20 to 60 light-days. This gives the radius of the BLR, so its intrinsic luminosity can be estimated. And knowing this makes it possible to estimate the actual distance from the observed luminosity.