IFRS: The first supermassive black holes?

Today’s Radio Galaxy Zoo post is by Ray Norris, our Project Advisor. Ray researches how galaxies formed and evolved after the Big Bang, using radio, infrared, and optical telescopes.

In Radio Galaxy Zoo, some bright radio sources don’t have any infrared sources at all associated with them, and they have been given the hashtag #ifrs, for Infrared-Faint Radio Sources. So what are these IFRS?

In 2006, we discovered about 1000 radio sources in the Australia Telescope Large Area Survey (ATLAS). Conventional wisdom told us that all of these would be visible in the infrared observations taken by the Spitzer Space Telescope, as part of the SWIRE project. So we were astonished to find that about 50 of our sources were not listed in the SWIRE catalog. Could they be bugs in our data? After eliminating iffy sources, we were left with 11 sources that are bright in the radio but invisible in the infrared images. We dubbed these objects Infrared-Faint Radio Source (IFRS). With hindsight, we should have thought of a better name for them. But at the time we didn’t know that they would turn out to be important!

A very deep infrared image (from this paper), made by stacking images from the Spitzer Space Telescope at the positions of IFRS (shown by the cross-hairs). This image, about 300 times deeper than the WISE images currently being used in Radio Galaxy Zoo, shows that not only is there no infrared counterpart at WISE levels, but even if you go enormously deeper, there’s still almost nothing there in the infrared.

We suggested in 2006 that these might be high-redshift active galaxies – galaxies whose emission is dominated by a super-massive black hole at their centre (the galaxies you are looking at in Radio Galaxy Zoo). This was surprising, because we were finding so many of them that it meant there must be far more supermassive black holes in the early Universe than found by deep optical surveys, such as SDSS (whose images we use in Galaxy Zoo). It’s also far more than can be accounted for by conventional hierarchical models of super-massive black hole formation. Naturally, our colleagues were sceptical, and most of us harboured our private doubts too. But over the next few years we tested this idea and its alternatives. Gradually our confidence grew that these were indeed high-redshift active galaxies.

In 2011 we showed that they were similar to high-redshift radio galaxies (HzRG) but even more extreme. Crucially, we suggested that they follow the same correlations between the radio and infrared emission as the HzRG. If this suggestion turned out to be correct, then that would push them to be amongst the first supermassive black holes in the Universe.

Fortunately, it was possible to test this hypothesis by measuring the redshifts of less extreme objects, to see if they followed these same correlations. Two new papers confirm that they do indeed follow this correlation. In one, Andreas Herzog and his colleagues use the European Very Large Telescope to measure the redshifts of three of these less extreme objects, and find they lie on the correlation, at redshifts between 2 and 3, just as predicted. In the other paper Jordan Collier and his colleagues take exactly the same data now being shown in Radio Galaxy Zoo, and search for objects which are relatively much brighter in the FIRST (radio) data than in the WISE (infrared) data. 1,317 of these are found, of which 19 have measured redshifts. Again, all but one of these lie in the redshift range 2 to 3. This is strong support for the hypothesis!

Armed with this, we are increasingly confident that the most extreme IFRSs that will turn up in the fainter ATLAS and COSMOS field, to be released in Radio Galaxy Zoo in a few weeks, will include many supermassive black holes formed in the first half-billion years after the Big Bang. According to conventional hierarchical black hole formation models, these shouldn’t exist. So the race will be on to identify them and measure their redshifts using instruments like ALMA.

Not everything in Radio Galaxy Zoo classified as being an IFRS will turn out to be a high-redshift black hole, as the data currently being displayed (FIRST and WISE) are not deep enough to pick out the really high-redshift objects. But when the new ATLAS data are loaded into RGZ in a few weeks, almost every object that appears in the radio but not in the infrared will be one of these enigmatic objects. We can’t wait to see how many you find!