Is there really a dichotomy in AGN jet power?

To gain new insights into the radio-loud/radio-quiet dichotomy reported for active galactic nuclei, we examine radio loudness as a function of Eddington ratio for a previously published sample of 199 AGN from five different populations. After initially considering radio loudnesses derived using total radio luminosities, we repeat the investigation using core radio luminosities only, applying a previously established mass correction for these core luminosities. In both cases, for Eddington ratios < 1 per cent, Fanaroff-Riley type I and broad-line radio galaxies are on average more radio-loud than Seyfert and low-ionization nuclear emission-line region galaxies. However, the distribution of radio loudnesses for the mass-corrected, core-only sample is much narrower than that of the clearly bimodal total radio loudness distribution. The advantages and disadvantages of using core- or lobe-dominated radio luminosity as a measure of instantaneous jet power are discussed. We furthermore compare the core and total radio luminosities for the entire sample, as well as illustrating the importance of the mass term by comparing the AGN with a sample of black hole X-ray binaries. We conclude that if the mass-corrected core radio luminosity is a good measure of jet power, then black hole spin may have considerably less impact on jet power than previously reported, or that our sample does not include the extremes of spin. If the spread in jet power is small then we suggest that characteristics of the ambient environment and/or the radio source age could be equally as important in producing a radio-loud/radio-quiet dichotomy seen in total radio luminosity.

💡 Research Summary

This paper revisits the widely discussed radio‑loud/radio‑quiet dichotomy in active galactic nuclei (AGN) by re‑examining the 199‑object sample originally analyzed by Sikora, Stawarz & Lasota (2007, hereafter SSL07). SSL07 used the total 5 GHz radio luminosity (core + extended lobes) together with the optical B‑band luminosity to define a radio‑loudness parameter R = L₅/L_B, and plotted R against the Eddington ratio λ = L_bol/L_Edd. They found two roughly parallel tracks: a “radio‑loud” sequence populated by broad‑line radio galaxies (BLRGs), radio‑loud quasars (RLQs) and FR I radio galaxies, and a “radio‑quiet” sequence occupied by Seyferts, LINERs and Palomar‑Green quasars. The separation was most pronounced for λ < 10⁻² (i.e., below 1 % of the Eddington limit). SSL07 interpreted the bifurcation as a manifestation of black‑hole (BH) spin: high‑spin BHs in massive ellipticals produce powerful jets, while low‑spin BHs in spirals generate weak jets.

The present study asks whether this dichotomy persists when (i) only the compact, core radio emission is used—thereby probing the instantaneous jet power rather than the integrated, often aged, lobe emission—and (ii) a mass‑correction term, motivated by the “fundamental plane” of BH activity, is applied to the radio‑loudness. The mass correction follows from the empirical scaling L₅,core ∝ L_X M^{0.8} (Merloni, Heinz & Di Matteo 2003) and the assumption that the optical B‑band luminosity is proportional to the X‑ray luminosity. Substituting L_B for L_X yields R ∝ λ^{‑0.4} M^{0.4}, or equivalently log R = ‑0.4 log λ + 0.4 log M + constant. Thus, after correcting for BH mass, the radio‑loudness becomes R′ = R · M^{‑0.4}.

Core radio flux densities at 5 GHz were collected from the literature (VLA, ATCA, VLBI, etc.) for each source, corrected for redshift and spectral index (α ≈ 0 when unknown). For 34 objects only upper limits on the core flux were available; these were treated as detections for the purpose of statistical tests, with the caveat that the true distribution may be broader. BH masses were taken directly from SSL07, which employed a variety of methods (reverberation mapping, stellar velocity dispersion, etc.) and have typical uncertainties of ~0.5 dex.

Three diagnostic plots were produced: (i) the original SSL07 total‑radio R versus λ, showing a clear bimodal distribution; (ii) core‑only R versus λ, where the two sequences remain but the gap narrows dramatically; (iii) mass‑corrected core R′ versus λ, where the two sequences essentially merge into a single, relatively narrow distribution. Two‑dimensional Kolmogorov–Smirnov (KS) tests were applied to the FR I/BLRG (radio‑loud) and Seyfert/LINER (radio‑quiet) subsamples. In case (i) the KS probability that the two groups are drawn from the same parent distribution is <10⁻⁶, confirming a strong statistical separation. In case (ii) the probability rises to ~10⁻³, indicating a weaker but still significant difference. In case (iii) the probability further increases to ~10⁻², suggesting that after mass correction the distinction is marginal.

These results have several important implications:

1. Core vs. total radio emission – The total radio luminosity includes contributions from large‑scale lobes that can persist for tens of Myr after the central engine has switched off or weakened. Consequently, total R may over‑estimate the instantaneous jet power. Core R, by contrast, reflects the current state of the jet base and is therefore a more direct probe of the instantaneous jet power.

2. Mass dependence – The mass‑corrected core radio loudness collapses the two sequences, implying that BH mass is a dominant factor in setting the radio output at a given λ. This agrees with theoretical expectations that jet power scales with BH mass because more massive BHs can anchor larger magnetic fluxes and have larger jet cross‑sections.

3. Spin contribution – If the mass‑corrected core radio luminosity truly traces jet power, the modest spread (≈0.7 dex in jet power for a given λ) suggests that variations in BH spin contribute less to the observed radio loudness than previously thought, at least within the sampled population. It is possible that the sample does not contain the most extreme spin values, or that spin‑driven power variations are masked by environmental effects.

4. Environmental and age effects – The authors argue that differences in ambient medium density, jet confinement, and source age can also modulate the observed radio luminosity, especially for the extended lobes. Thus, the classic radio‑loud/radio‑quiet dichotomy seen in total radio power may be largely driven by extrinsic factors rather than intrinsic jet physics.

5. AGN–X‑ray binary scaling – By comparing the AGN data with a sample of black‑hole X‑ray binaries (BHBs), the authors show that the mass‑corrected core radio–X‑ray plane aligns with the established “fundamental plane” of BH activity, reinforcing the universality of the mass scaling across many orders of magnitude in BH mass.

In conclusion, the apparent radio‑loud/radio‑quiet dichotomy in AGN is strongly dependent on the choice of radio metric. When using total radio luminosities, a clear bimodality emerges, supporting earlier claims of a spin‑driven division. However, when focusing on core radio emission and correcting for BH mass, the distribution becomes narrow and essentially unimodal. This points to a scenario where jet power is primarily governed by BH mass and external conditions (environment, age), with spin playing a secondary role, if any. Future work should aim for larger, uniformly selected samples with high‑resolution core radio measurements, better constraints on BH spin (e.g., via Fe Kα line modeling or continuum fitting), and multi‑wavelength diagnostics that can disentangle age and environmental effects. Only then can we definitively assess whether a true intrinsic dichotomy exists in AGN jet power.