A team of KIPAC astrophysicists has applied a rigorous statistical analysis to observations of quasars resulting in an interesting perspective.
An example of a bias arising from data truncation. In this plot of radio luminosity versus redshift (distance) for quasars detected by a survey, inherently faint objects can only be seen if they are close (low redshift).
Quasars are distant galaxies that appear bright to us in every kind of light, from radio to optical to x-rays. They outshine most of their neighbors because it isn't just the stars and interstellar matter that glows in quasars, but also an active center where jets of particles and radiation originate from near a supermassive black hole. These 'active' galaxies, known to astronomers as "AGN", present some of the most interesting enigmas in the Universe.
One question that has persisted since almost their first discovery 40 years ago is whether there are distinct sub-populations of quasars. A property that is often of interest in this regard is the "radio loudness" which is the ratio of the radio to optical luminosity of the quasar at selected frequencies. Since the optical emission largely arises from the matter orbiting the central black hole while the radio emission is largely due to the jets, the radio loudness has been taken as an indicator of the relative strength of the jet. For more than 30 years there have been conflicting indications of whether quasars form two populations - so-called "radio loud" and "radio quiet" divided by a radio loudness of around ten - or whether this division is arbitrary and there is a continuum of radio loudness among quasars.
KIPAC professor Vahe Petrosian, postdoc Jack Singal, former postdoc Lukasz Stawarz, and affiliate Andrew Lawrence have undertaken a new analysis of observational data from quasars to address this and other questions. The analysis features statistically rigorous techniques developed by Petrosian to deal with the biases and correlations inherent in the data that often skew the conclusions. For instance, as an example of bias, at farther distances - which also means farther back in cosmic time - only the most luminous objects will be seen in a given survey, while at closer distances less luminous objects can be seen as well. As an example of a correlation, the optical and radio luminosities are not independent, such that one cannot attempt to determine how, as a group, quasars' radio and optical emission have evolved in the Universe over time by considering them separately.
The team's methods have produced some intriguing results. They have shown that with the data they use, a less rigorous analysis produces a distribution of radio loudnesses that hints at there being two populations divided by a radio loudness of around ten, while in the more rigorous analysis that indication disappears. This seems to heavily favor the 'one population' explanation. They also disentangle the optical and radio evolution of quasars as a group over time, and find that in the past the radio luminosity was somewhat enhanced over the optical luminosity relative to the present. Lastly, the analysis shows that the radio and optical luminosities are correlated such that the more luminous quasars are also more radio loud.
All of these results have implications for astrophysics and cosmology. If quasars are truly a continuum in regard to radio loudness, then there would seem to be no critical difference that could arise in the properties of the central black hole and jet launching systems, only a smooth spectrum of properties. However, the enhancement of the radio emission relative to optical at early times could be related to some global property of the systems that changes in bulk over time. The KIPAC team next plans to use their methods to perform a similar analysis of data from observations of Blazars with the Fermi gamma-ray space telescope.
This work is based on a paper submitted to the Astrophysical Journal and available from astro-ph at arXiv:1101.2930.