Title: Hyperluminous Starburst gives up its secrets

Authors: R. J. Ivison, M. J. Page, M. Cirasuolo, C. M. Harrison, V. Mainieri, V. Arumugam and U. Dudzeviciute

First author institution: European Southern Observatory, Karl-Schwarzschild-Strasse 2, D-85748 Garching, Germany

Status: published in MNRAS, open access

Surveys of the infrared sky have led to the discovery of thousands of dust obscured, highly star-forming galaxies — often referred to as submillimetre galaxies (SMGs) due to the submillimetre-wave emission that characterises their cool, dusty nature. Unfortunately, this long wavelength emission presents an observational challenge: the resolution of a telescope has a physical limit, directly proportional to the wavelength of light divided by the diameter of the telescope. For longer wavelengths such as submillimetre, a telescope must be much larger than an optical telescope to achieve comparable resolution. As such, submillimetre telescopes are limited by the feasible sizes of single mirrors, and for decades infrared astronomy was stuck with low resolution images. Out of which has grown a science of fuzzy blobs* (see Fig 1).

Since the advent of ALMA, the Atacama Large Millimetre Array, these infrared-bright sources have been under scrutiny at much higher resolution made possible by the power of interferometry**. Upon high res inspection, many of these very bright sources turn out to be the combined light of several galaxies, merged together in a lower resolution telescope observation. This paper investigates one such source, and finds some very curious things indeed about the nature of this particular fuzzy blob.

This submillimeter source (lengthily named HATLAS J084933.4+021443) was identified as a dusty starburst galaxy in a sky survey with the Herschel space telescope, but when observed with ALMA breaks up into five galaxies labelled W, C, M, T and E. One of these — W — is so bright in the submillimetre that it classifies as a rare hyperluminous infrared galaxy (HyLIRG), implying a star formation rate of many thousands of stars per year, orders of magnitude higher than your average galaxy! HyLIRGs offer a chance to examine the physical mechanisms for star formation at its most extreme: what is it that makes these galaxies form stars at such a rate? Do they represent an important phase of star formation in the universe? And how does star formation in galaxies relate to the growth of the supermassive black holes that may be lurking at their centres, masked by huge reservoirs of dust? As the authors put it, “hidden quasar, or protogalaxy? Monster, or baby?”

The authors of this paper gather observations across the spectrum to investigate the physics of this group of galaxies — and in particular, galaxy W, the brightest of the bunch. From optical and X-ray spectroscopy, they determine that galaxy W does indeed host an active galactic nucleus (AGN): that is, a supermassive black hole that is actively feeding. There is still some speculation as to whether all HyLIRGs host AGN, as it is notoriously challenging to observe an AGN embedded in a galaxy thick with dust, obscuring potential AGN signatures from the cores. Galaxy W, surprisingly, appears to be relatively unobscured.

But the most puzzling aspect of galaxy W observations reveals itself in the UV to radio spectral energy distribution (SED, see figure 2): galaxy W looks like an average star forming galaxy. In an SMG, even if much of the AGN radiation is absorbed by dust, that dust re-radiates the energy in the mid infrared creating a distinctive clue as to the powerful engine hidden inside. But galaxy W has no such signature. So a conflict arises: this galaxy must host a powerful AGN due to its high X-ray luminosity, which cannot be explained any other way, but it shows absolutely no indication of doing so in the rest of the spectrum.

The authors suggest three possible scenarios that could give rise to this confusing combination of observations. First, they suggest that we are viewing the central AGN in the galaxy through a hole in its dusty exterior. Since the AGN is so incredibly powerful, it is plausible that it could fairly quickly blast out a cavity – and with the right alignment, the central region would become exposed.

Secondly, they suggest that there may be two AGN here: one, hidden deep in the dusty galaxy, and a second exposed AGN associated with a second galaxy perhaps left over from a merger event. This is also a nice solution, as a recent merger could have triggered a starburst, and we would be looking at two galaxies at two distinct evolutionary stages: that of a young dusty AGN with a high star formation rate, and an evolved neighbour with an unobscured AGN.

The third and perhaps most unusual scenario is that of an ejected AGN. Given the right combination of mass ratios and black hole spins, it is possible for a galaxy merger to result in the SMBH being flung out of the galaxy at high velocity. If this is the case, the AGN can continue to shine for tens of millions of years. In this scenario, galaxy W continues to form stars, without the presence of an AGN which could regulate its star formation, giving rise to the unusually high star formation rate & infrared luminosity that classes it as a HyLIRG. The jury is still out on the role of AGN in regulating star formation, but this could be a rare exception in the sequence of galaxy evolution.

The authors conclude that, while an AGN is certainly present in the system, there is more to galaxy W than meets the eye, and further investigation — such as deeper X-ray observation — is necessary to make any more conclusive claims about its nature. Nevertheless, these observations add to the growing consensus that HyLIRGs in general host AGN, and that the interplay between AGN activity and star formation is still a very relevant topic in questions of galaxy evolution.

*“The science of fuzzy blobs”: current contender for the title of my thesis

** Every ALMA paper I read blows me away – I think this telescope is just amazing