When astronomers add up all the gas and dust contained in ordinary galaxies (like our own Milky Way), they find a discrepancy: there is not nearly enough matter for stars to form at the observed rates for long. As a (partial) solution, a matter cycle on gigantic scales has been proposed. In our local galactic neighbourhood, traces of this mechanism had already been found. Now, a study led by Kate Rubin of the Max Planck Institute for Astronomy has found the first direct evidence of such gas flowing back into distant galaxies that are actively forming new stars, validating a key part of "galactic recycling".
Figure 1: Images of the six galaxies with detected inflows taken with the Advanced Camera for Surveys on the Hubble Space Telescope. Most of these galaxies have a disk-like, spiral structure, similar to that of the Milky Way. Star formation activity occurring in small knots is evident in several of the galaxies' spiral arms. Because the spirals appear tilted in the images, Rubin et al. concluded that we are viewing them from the side, rather than face-on. This orientation meshes well with a scenario of 'galactic recycling' in which gas is blown out of a galaxy perpendicular to its disk, and then falls back in at different locations along the edge of the disk.
Star formation regions, such as the Orion nebula, create some of the most beautiful astronomical sights. It is estimated that in our home galaxy, the Milky Way, on average one solar mass's worth of matter per year is turned into stars. Yet a survey of the available raw material, clouds of gas and dust, shows that, using only its own resources, our galaxy could not keep up this rate of star formation for longer than a couple of billion years. Is our home galaxy currently undergoing a rather special, short-lived era of star formation? Both stellar age determinations and comparison with other spiral galaxies show that not to be the case. One solar mass per year is a typical star formation rate, and the problem of insufficient raw matter appears to be universal as well.
Evidently, additional matter finds its way into galaxies. One possibility is an inflow from huge low-density gas reservoirs filling the intergalactic voids; there is, however, very little evidence that this is happening. Another possibility, closer to home, involves a gigantic cosmic matter cycle. Gas is observed to flow away from many galaxies, and may be pushed by several different mechanisms, including violent supernova explosions (which are how massive stars end their lives), and the sheer pressure exerted by light emitted by bright stars on gas in their cosmic neighbourhood.
As this gas drifts away, it is pulled back by the galaxy's gravity, and could re-enter the same galaxy in time scales of one to several billion years. This process might solve the mystery: the gas we find inside galaxies may only be about half of the raw material that ends up as fuel for star formation. Large amounts of gas are caught in transit, but will re-enter the galaxy in due time. Add up the galaxy's gas and the gas currently undergoing cosmic recycling, and there is a sufficient amount of raw matter to account for the observed rates of star formation.
There was, however, uncertainty about the viability of this proposal for cosmic recycling. Would such gas indeed fall back, or would it more likely reach the galaxy's escape velocity, flying ever further out into space, never to return? For local galaxies out to a few hundred million light-years in distance, there had indeed been studies showing evidence for inflows of previously-expelled gas. But what about more distant galaxies, where outflows are known to be much more powerful – would gravity still be sufficient to pull the gas back? If no, astronomers might have been forced to radically rethink their models for how star formation is fueled on galactic scales.
Now, a team of astronomers led by Kate Rubin (MPIA) has used the Keck I telescope on Mauna Kea, Hawai'i, to examine gas associated with a hundred galaxies at distances between 5 and 8 billion light-years (z ~ 0.5 – 1), finding, in six of those galaxies, the first direct evidence that gas adrift in intergalactic space does indeed flow back into star-forming galaxies. As the observed rate of inflow might well depend on a galaxy's orientation relative to the observer, and as Rubin and her team can only measure average gas motion, the real proportion of galaxies with this kind of inflow is likely to be higher than the 6% directly suggested by their data, and could be as high as 40%. This is a key piece of the puzzle and important evidence that cosmic recycling ("galactic fountains") could indeed solve the mystery of the missing raw matter.
The results described in this release have been published as Kate H. R. Rubin et al., "The Direct Detection of Cool, Metal-Enriched Gas Accretion onto Galaxies at z ~ 0.5" in the journal Astrophysical Journal Letters, Vol. 747 (2012), p. 26ff. The co-authors are Kate H. R. Rubin (Max Planck Institute for Astronomy), J. Xavier Prochaska (MPIA and UCO/Lick Observatory, University of California), David C. Koo (UCO/Lick Observatory), and Andrew C. Phillip (UCO/Lick Observatory).
Which telescopes/instruments were used in this study?
The data used in this study were obtained with the Low Resolution Imaging Spectrometer (LRIS) at the Keck I telescope on the summit of Mauna Kea, Hawai'i. Additional images indicating galaxy orientation were obtained using the Advanced Camera for Surveys (ACS) on the Hubble Space Telescope. The Keck telescope is operated by the W. M. Keck observatory. The observatory is governed by the California Association for Research in Astronomy (CARA), whose Board of Directors includes representatives from the California Institute of Technology and the University of California, with liaisons from NASA and the Keck Foundation. The Hubble Space Telescope is a project of international cooperation between the US space agency NASA and the European Space Agency ESA.
How were the inflows measured?
Using the so-called Doppler effect, spectroscopic measurements can be used to determine a gas cloud's velocity directly towards, or away from, an observer. For galaxies as distant as these, individual gas clouds are impossible to resolve. Instead, Rubin and her colleagues used a simple model positing the presence of two clouds: one at rest relative to the galaxy's stars, one moving. Fitting this model to the data, they obtained an average velocity for the moving gas. In 2/3 of the cases, this averaged motion was outwards. In six cases, motion was inwards, back towards the galaxy.
How do we know this gas originally came from the galaxy in question?
The detection method uses spectral lines associated with the chemical elements magnesium and iron. These elements are produced in stars, and are not present in the pristine intergalactic medium. Hence, measurements of this kind cannot detect inflow onto a galaxy of pristine intergalactic matter (although astronomers are very actively searching for this kind of inflow!). Instead, the gas clouds in question are clearly galactic matter; and as the observed galaxies are isolated, matter previously ejected by the galaxy itself is the most likely candidate. There is, however, an outside chance that the gas belongs or once belonged to small dwarf galaxies being attracted to their larger cousins.
Why is 6% a lower limit only?
In- and outflow probably occur in certain preferred directions. Conspicuously, we see five of those six galaxies pretty much from the side (rather than face-on). This could be explained if outflow occured predominantly perpendicular to the disc, and inflow predominantly sideways. In such a scenario, we would only detect average inflow in galaxies we see oriented with their edges toward us. Quite generally, the method used only gives an average value for the motion of the gas. Thus, weaker inflows can be masked by stronger outflows.
How does this improve upon previous results?
Rubin and her team made the first direct observation of gas inflow onto distant, actively star-forming spiral galaxies. In the local universe, inflow into spiral galaxies had been previously observed (Wakker 2001 and Lehner & Howk 2011 for the Milky Way; Sancisi et al. 2008 for galaxies with distances up to about two hundred million light years). For more distant galaxies, in which galactic winds - outflows of gas – are known to be much stronger, the data was inconclusive: previous observations of distant (redshift z > 0.5) star-forming galaxies (e.g., Weiner et al. 2009, Steidel et al. 2010) using the same method had only been able to detect outflows – which is readily explained by those studies' need to average over several hundred galaxies in order to be able to extract an average motion value from their data. In such averaging, the inflow signal from some of the observed galaxies will be drowned out by the much more common outflow signals. There have been previous observations finding inflows of gas (Sato et al. 2009, Coil et al. 2011), but in those cases, the objects under study were older galaxies in which star formation had already ceased. Rubin et al.'s work provides the first direct link between inflows and distant star-forming spiral galaxies – a key ingredient in finding the solution to the mystery of the missing raw matter.