[Paper Review] AGNs and Starbursts: What Is the Real Connection?

Luis C. Ho|arXiv (Cornell University)|Nov 6, 2005

Solar and Space Plasma Dynamics3 citations

TL;DR

This paper investigates the physical connection between active galactic nuclei (AGNs) and starbursts, finding strong observational evidence for a causal link through post-starburst populations in luminous AGNs. Despite abundant gas, star formation is suppressed in AGN hosts due to feedback, with far-infrared emission in quasars primarily powered by the AGN rather than stars, challenging assumptions about star formation rates derived from infrared luminosity.

ABSTRACT

It is now widely believed that the growth of massive black holes is closely linked to the formation of galaxies, but there have been few concrete constraints on the actual physical processes responsible for this coupling. Investigating the connection between AGN and starburst activity may offer some empirical guidance on this problem. I summarize previous observational searches for young stars in active galaxies, concluding that there is now compelling evidence for a significant post-starburst population in many luminous AGNs, and that a direct, causal link may exist between star formation and black hole accretion. Quantifying the ongoing star formation rate in AGNs, however, is much more challenging because of the strong contamination by the active nucleus. I discuss recent work attempting to measure the star formation rate in luminous AGNs and quasars. The exceptionally low level of coeval star formation found in these otherwise gas-rich systems suggests that the star formation efficiency in the host galaxies is suppressed in the presence of strong AGN feedback.

Motivation & Objective

• To determine whether AGN activity and star formation are physically linked or merely coincidental in luminous galaxies.
• To assess the role of AGN feedback in suppressing star formation efficiency in gas-rich host galaxies.
• To resolve the origin of far-infrared emission in quasars, distinguishing between heating by young stars versus AGN accretion.
• To investigate differences in star formation rates between Type 1 and Type 2 quasars, particularly regarding obscuration mechanisms.
• To quantify ongoing star formation in AGNs using [O II] emission as a tracer, overcoming contamination from the active nucleus.

Proposed method

• Analyzing optical absorption-line spectra of AGNs to detect A-type stars indicative of intermediate-age post-starburst populations.
• Using the Sloan Digital Sky Survey (SDSS) sample of Type 2 AGNs to correlate post-starburst fraction with AGN luminosity, establishing statistical significance.
• Measuring [O II] λ3727 emission line fluxes as a direct tracer of ongoing star formation rates (SFRs), correcting for AGN contamination.
• Comparing [O II]-derived SFRs with those inferred from far-infrared (FIR) luminosity to test whether FIR emission is powered by stars or the AGN.
• Examining composite spectra of SDSS Type 2 quasars to compare [O II]/[O III] line ratios with those of Type 1 quasars, revealing differences in star formation activity.
• Evaluating the impact of AGN feedback on molecular clouds by modeling ionization and thermal structure, and assessing suppression of star formation efficiency.

Experimental results

Research questions

• RQ1 Is there a direct causal link between star formation and AGN activity, or is the observed coexistence merely coincidental?
• RQ2 Why is star formation efficiency anomalously low in AGN hosts despite the presence of abundant molecular gas?
• RQ3 To what extent is the far-infrared emission in quasars powered by young stars versus the AGN's accretion energy?
• RQ4 Why do Type 2 quasars exhibit significantly higher star formation rates than Type 1 quasars, and what does this imply about obscuration mechanisms?
• RQ5 How does AGN feedback suppress star formation in host galaxies, and what physical mechanisms underlie the observed low SFRs?

Key findings

• There is compelling observational evidence for a significant post-starburst population in luminous AGNs, with the fraction increasing with AGN luminosity, indicating a causal link.
• Ongoing star formation rates in Type 1 quasars are very low—on the order of a few solar masses per year—despite the presence of abundant molecular gas.
• The [O II]-derived SFRs in the PG quasar sample are at least an order of magnitude lower than those inferred from FIR luminosity, indicating that the AGN, not stars, powers most of the far-infrared emission.
• Type 2 quasars exhibit a [O II]/[O III] line ratio approximately an order of magnitude stronger than Type 1 quasars, implying a higher SFR of ~20 M☉ yr⁻¹, comparable to a local starburst.
• The low star formation efficiency in AGN hosts suggests that AGN feedback suppresses star formation, even in gas-rich environments.
• The data challenge the assumption that FIR emission traces star formation, showing that AGN heating can mimic stellar heating in dust emission.