Reconstructing the cosmic history of star formation: ALMA takes stock of the fuel for star formation in distant galaxies

22. September 2016
A study by a large international team led by Fabian Walter of the Max Planck Institute for Astronomy, using the millimeter telescope ALMA, has traced the raw building blocks of star formation back in time to an era about 2 billion years after the big bang, yielding clues as to the history of star formation in our universe. The study targeted one of the best studied regions of the sky: the Hubble Ultra Deep Field (HUDF), first imaged in depth by the Hubble Space Telescope in 2003/2004. This is the first time a millimeter wave image that includes spectral information has been taken of this portion of the HUDF, sufficient to show galaxies whose light took up to 11 billion light years to reach us. 

In-depth information: Reconstructing the cosmic history of star formation: ALMA takes stock of the fuel for star formation in distant galaxies

Astronomical objects look less bright with increasing distance. The most striking case is that of our Sun: The Sun is a rather average star. The only reason why it looks so much brighter than the stars we see in the night sky is its proximity; the distance to the nearest star after that is more than 60 000 times Earth's distance to the Sun. This defines one of the major strands of astronomical progress: with ever larger telescopes collecting ever more light in one go, and with long exposure times, astronomers have obtained more and more images of fainter and fainter objects. In astronomy, every observation of distant objects is an observation of our cosmic past - we see the Andromeda galaxy not as it is now, but as it was 2.5 million years ago, since it takes that long for the galaxy's light to reach us. That is why the quest for fainter and fainter objects, for ever more distant targets, is also a quest for peering ever deeper into our universe's past.

Hubble Deep Field and Ultra Deep Field

A milestone of this quest was the Hubble Deep Field: a total of 140 hours of exposure with the Hubble Space Telescope, aimed at a small region of the Northern sky, a mere 1 percent of the area covered by the full moon, that was virtually free of stars and obscuring dust, and allowed an excellent view of around 3000 distant galaxies. The nearly 1000 scientific articles that have referenced the Hubble Deep Field data are a vivid demonstration of the value of this kind of deep, representative cosmic sample. In 2003/2004, an even deeper image data was taken for the Hubble Ultra Deep Field, with the same size in the sky, and containing almost 10 000 galaxies, the most distant of them so far away that their light has been travelling for about 13 billion years before reaching us. Crucially, the Hubble Ultra Deep Field is visible to major telescopes in the Southern Hemisphere, notably the telescopes of the European Southern Observatory in Chile and, more recently, the nearby international submillimeter/millimeter observatory ALMA ("Atacama Large Millimeter/submillimeter Array") in the Atacama desert.

Astronomers make use of various wavelength ranges of electromagnetic radiation to fully understand the objects they study. After the Hubble Space Telescope had delivered superb optical, near-infrared and ultraviolet images of the Ultra Deep Field, which is particularly suited for showing the light of the stars contained in these galaxies, it did not take long for other observatories to supply observations in other wavelengths. X-ray observations, for instance, allow astronomers to study active galaxies (galaxies hosting accreting supermassive black holes) in the sample.

Supplying the missing millimeter wavelength

But until now, one important wavelength region was missing in the effort to chart remote cosmic history: millimeter and submillimeter radiation. Until the ALMA observatory, a telescope compound using 66 antennas that can be combined to act as a single, large telescope, reached its full strength in 2015, millimeter/submillimeter observations of the most interesting galaxies in the Ultra Deep Field were beyond the existing technology. Those galaxies include numerous distant galaxies that we see as they were a mere 3 billion years after the Big Bang, nearly 11 billion years before our time (redshift z=2), and their submillimeter observations would allow astronomers a peek into an especially interesting epoch in cosmic history.

At present, astronomers only have a broad-brush picture of cosmic history, but that picture already includes some curious long-term trends. One large-scale property of various cosmic epochs is their overall rate of star formation. Stars are formed when molecular gas within galaxies collapses under its own gravity, and certain regions within heat up sufficiently for nuclear fusion to set in. Our Milky Way is currently forming stars at a modest rate of about 1 solar mass per year. But studies of star formation in distant galaxies show that galaxies in the early universe were much more productive. Typically, such studies examine spectral lines whose light is emitted when molecular clouds collapse and heat up in the process. By this criterion, looking at an overall average for forming new stars, star formation productivity increased since the Big Bang, reaching a peak of maximum productivity about 3 and 6 billion years after the Big Bang, and has been declining ever since. The present universe is producing stars at a rate that is only about one tenth of this maximum productivity.

One of the key tasks of extragalactic astronomy is to understand this development: Was it a matter of supply and demand, with new reservoirs of molecular gas becoming available over time? What was the role of heating mechanisms, such as supernovae, large numbers of young hot stars, or active galactic nuclei?

Using carbon monoxide to track hydrogen molecules

Millimeter and submillimeter observations of distant galaxies can supply an important part of the puzzle. Where optical observations, such as with the Hubble Space Telescope, show the presence of atoms and their characteristic spectral lines, infrared observations typically show spectral lines associated with molecules, and with the thermal radiation of dust. For distant galaxies, whose light is redshifted by cosmic expansion, such molecular lines are shifted into the millimeter radiation regime.

Of special interest for star formation historians are spectral lines associated with carbon monoxide, CO. Studies both in our Milky Way and targeting isolated distant galaxies show that the ratio between hydrogen molecules - as the raw material for star formation - and carbon monoxide molecules has largely been constant for all normal galaxies for the last 12 or so billion years. In this way, it is possible to use millimeter observations of CO to deduce the amount of molecular hydrogen that is present. With such measurements, a history of star formation rates can be supplemented by a history of star formation's raw material - a key ingredient for understanding what has been going on in these galaxies.

A 3D map and cross-checks

As those observations make use of spectral lines, which form a characteristic pattern at clearly defined frequencies, they also allow for the determination of the redshift of the matter clouds observed: the shift of radiation towards lower frequencies that is a direct consequence of cosmic expansion. Using the standard cosmological models, the cosmic redshift of spectral lines allows astronomers to infer the distance of the object in question, allowing for the creation of 3D maps of the cosmos.

Continuum radiation, which is not confined to specific frequencies, yields further information. Overwhelmingly, this radiation is thermal radiation from cosmic dust. First of all, there is a direct relationship between the amount of dust and the amount of molecular hydrogen present in a galaxy. While this relation is not as tight and well-tested as the correlation between molecular hydrogen and carbon monoxide, it does allow for a consistency check. Also, the major source heating up cosmic dust within a galaxy is star formation, more precisely the radiation from hot, young, massive stars: From the thermal radiation, and hence the temperature of the dust, it is possible to deduce a galaxy's star formation rate.

Up until now, such millimeter/submillimeter measurements were only possible for single, isolated galaxies targeted directly. When Fabian Walter at the Max Planck Institute for Astronomy and his international colleagues considered the capabilities of ALMA, they realized that the observatory's sensitivity and resolution (capability for discerning small details) might make it possible to conduct a fruitful "blind search" in the Ultra Deep Field, complementing the many existing observations of that region of the sky.

40 hours of ALMA in the Ultra Deep Field

Between July 2014 and April 2015, ALMA spent a total of 40 hours with 50 of its antennas pointed at the Southern constellation of Fornax, where the Ultra Deep Field is located, catching light from an area of the sky one arc minute times one arc minute, about 1/7 of the Hubble Ultra Deep Field. Shortly after the observations were concluded, Walter and his colleagues knew that their blind search had been successful. A more thorough analysis over the following months confirmed the detection of ten galaxies. The researchers present their results today at the Half a Decade of ALMA conference in Palm Springs, California. The results also are accepted for publication in a series of seven scientific papers appearing in The Astrophysical Journal. 

The observations show an intriguing picture of the raw material of star formation during the "Golden Age" of our universe's star formation maximum. “The new ALMA results imply a rapidly rising gas content in galaxies with increasing look-back time,” says Manuel Aravena, an astronomer with the Universidad Diego Portales in Santiago, Chile, and Co-leader of the research team. “This increasing gas content is likely the root cause for the remarkable increase in star formation rates during the peak epoch of galaxy formation, some 10 billion years ago.”

The galaxies that were detected with these millimeter observations can also be identified in the optical and infrared observations of the deep field - although no-one could have predicted precisely which of the galaxies visible in Hubble's iconic image would shine brightly in the milimeter regime. In fact, while the sample is still small, there are indications that the current brightness of a galaxy's stars and its molecular hydrogen content (that is, the reservoir to form future stars) are less closely related than some researchers had postulated. As Chris Carilli, an astronomer with the National Radio Astronomy Observatory (NRAO) in Socorro, New Mexico, and member of the research team says “[We have] discovered a population of galaxies that is not clearly evident in any other deep surveys of the sky.”

Paving the way for a Large Program

While these results are interesting in themselves, one of their most important function is as a feasibility study for additional large-scale observations. “These newly detected carbon-monoxide rich galaxies represent a substantial contribution to the star-formation history of the universe,” says Roberto Decarli, an astronomer with the Max Planck Institute for Astronomy (MPIA) in Heidelberg, Germany, and member of the research team. “With ALMA, we have opened a new pathway for studying the early formation and assembly of galaxies in the HUDF.”

In fact, based in part on the results they have now published, a large international team led by Walter, Aravena and Carilli have recently been rewarded the first Large Program with ALMA: An unprecedented 150 hours of exposure time that will cover essentially the entire Ultra Deep Field in 2017. Going by their present results, the astronomers expect this larger survey to yield on the order of 50 galaxies from redshift 0 (today) to when the Universe was only a few billion years old, allowing for some degree of statistics of star formation throughout cosmic history. Fabian Walter says: "We have yet to understand the exact causes of the cosmic star formation history. By supplementing the missing star-forming material, the approved ALMA Large Program will complete our view of the well-known galaxies in the iconic HUDF. As such, our Large Program will provide some key missing pieces of the puzzle of cosmic star formation".

[Back to top]

Zur Redakteursansicht