How galaxies have produced their stars: ASPECS survey provides key chapter of cosmic history

September 24, 2020

Astronomers have used the ALMA observatory to trace the fuel for star formation – molecular hydrogen gas – in the iconic Hubble Ultra-Deep Field, one of the best-studied regions of the sky. The observations allowed a group led by Fabian Walter of the Max Planck Institute for Astronomy to track how the universe’s inventories of gas and dust have changed over time from just two billion years after the big bang to the present. Comparing their own observations with additional observational data and modern simulations, the astronomers were able to characterize and quantify the gas flows that are necessary prerequisites for the formation of stars within galaxies. The result is a broad-brush history of cosmic star formation that includes all the important pieces: the history of star production itself as well as information about the supply chain that enables stars to be produced in the first place.

The supply chain for star production

Tracing the origin of a common household item, like an appliance, amounts to reconstructing a supply chain: the raw materials transformed into more elaborate components, and those components assembled into a finished product. If supplies are missing, production will slow down, or might even grind to a halt. Documenting the factory's inventory of the necessary components or raw material is a useful way of learning about the production history. 

When galaxies form stars, there is of course no planning behind it, economic or otherwise. Stars form whenever the conditions are right for them to form, whenever the right material is available. In order to produce stars, we need cool gas made of hydrogen molecules. Such cool gas is produced when a sufficiently dense cloud of warmer gas made of hydrogen atoms cools down – under the right conditions, the hydrogen atoms pair off, each pair forming a hydrogen molecule H2

The atomic hydrogen inventory can be replenished as well. There is a huge reservoir of ionized hydrogen in the vast spaces between galaxies, warm intergalactic plasma that contains more than 90% of all hydrogen in the universe. Keep track of how those inventories change over time, reconstruct the supply chain, and you can learn about the production history of stars. Keeping track of change is possible because astronomers always look into the past.

A deep look into cosmic history

If we point our telescopes at one of our nearest neighbors, the Andromeda galaxy M31, we see that galaxy as it was 2.5 million years ago, because it took the light we receive now 2.5 million years to travel from Andromeda to us. We cannot observe our own past that way, but we can do the next best thing: All our current knowledge points towards the fact that, on average, the universe is the same everywhere. Regardless of where in the cosmos we are: If we consider a suitably large region, at the present time, we will always find about the same number of larger galaxies, the same number of smaller galaxies, roughly the same number of stars, and the same amount of molecular gas.

That allows astronomers to reconstruct a cross-section of cosmic history. If you want to know what the average properties of the universe were, say, a billion years ago, look at objects so distant that their light takes a billion years to reach us! Repeat the process for different distances, corresponding to different cosmic epochs, and you will obtain at least an average history of the cosmos. The details will vary, but the big picture of cosmic evolution obtained in this way should be valid universally, providing clues about our own cosmic history over the past billions of years.

The history of stellar production rates

Over the past two decades, deep sky surveys using visible light and infrared radiation have given us a fairly complete picture of how many stars there were in galaxies in each cosmic epoch, from the first billion years after the big bang to the present. Particularly important was the Hubble Ultra-Deep Field (UDF): a small region in the sky, about one tenth the apparent diameter of the full moon, where the Hubble Space Telescope captured hundreds of images between 2003 and 2004, with a total of nearly 16 days exposure time, which were then combined into a single image. 

The UDF and other surveys lead to a consistent picture of star formation history, with star production ramping up to a veritable boom some 10 billion years ago, followed by a continuous decline in production rates. Half the stars in the universe had already been produced by the time the universe was 4.5 billion years old, a third of its current age. But why the increase and decline? To answer that, it makes sense to see how much raw material, molecular hydrogen, was available at different times. 

Molecular gas: the missing piece of the puzzle

This is where ASPECS comes in, the ALMA Spectroscopic Survey in the Hubble Ultra-Deep Field, organised by Fabian Walter (MPIA) and his colleagues. The astronomers used the ALMA observatory in Chile, fully operational since 2013, which can combine up to 50 large (sub)millimeter telescopes in what is called interferometry: a technique that combines telescopes in a way that allows the imaging of fine details that would only be accessible to a much larger single telescope. 

For studying molecular gas in distant galaxies, facilities like ALMA are ideal. Detecting cosmic molecules requires measuring light at specific wavelengths. Because our universe is expanding, there is what is known as the cosmological redshift: The more distant a galaxy is, the farther its light is shifted towards longer wavelengths. For distant galaxies, the wavelengths needed to deduce the presence of hydrogen molecules fall into the millimeter region of the electromagnetic spectrum, corresponding to short radio waves – which is exactly what ALMA was designed to observe. 

The overall collecting area of ALMA is much larger than for any previous millimetre/submillimetre telescopes, so the observatory is very sensitive. That is necessary, as the light reaching us from galaxies billions of light-years away is exceedingly faint. Before ALMA, a survey with the sensitivity of ASPECS would not have been possible. Even with ALMA, ASPECS needed a total of almost 200 hours of observation time, which makes it one of ALMA's so-called large programs – the first such program specifically searching for molecular gas in the distant universe.

An unbiased view of Hubble Ultra-Deep Field

In order to yield information that can be generalized to the universe as a whole, a survey such as ASPECS needs to be unbiased. (Consider the analogous situation of an opinion poll: In order to reconstruct public opinion, you will need a representative sample of respondents.) To that end, ASPECS chose the best-studied region of the sky, at least when it comes to distant galaxies: the Hubble Ultra-Deep Field (UDF). The combined image Hubble Ultra-Deep Field contains around 10,000 identifiable galaxies. Light from the most distant galaxy took 13 billion years to reach us. (For comparison: The big bang happened 13.8 billion years ago.) ASPECS scanned the Hubble Ultra-Deep Field at wavelengths around 1.3 mm and 3 mm. In their survey, the researchers followed an observational approach that had been shown to work well through a number of pilot programs, both with the IRAM Plateau de Bure Interferometer and with earlier ALMA observations. At those specific wavelengths, the Earth's atmosphere is virtually transparent, in particular at high-elevation locations such as the Chajnantor plateau in Chile where ALMA is located, at an elevation of 5000 meters.

More specifically, at each location within the Hubble Ultra-Deep Field, the astronomers took two spectra, carefully mapping the intensity of light received at different wavelengths between 1.1 and 1.4 mm, and also between 2.6 and 3.6 mm. In such spectra, molecules reveal themselves via so-called emission lines – narrow wavelength regions where there is a sharp maximum of intensity. While molecular hydrogen has no detectable emission lines, a molecule that is typically found in its company does: Carbon monoxide CO has a number of clearly detectable lines. 

From the nearby cosmos, we know that in a typical interstellar gas cloud, for each CO molecule, you will find on the order of 10,000 hydrogen molecules. As hydrogen molecules bump into CO molecules, the CO molecules gain energy – which they then emit in the form of electromagnetic radiation, at the wavelengths corresponding to their emission line. Measure the intensity of those CO lines, and you can deduce the amount of molecular hydrogen that is around in that specific region, occasionally bumping into CO. By taking into account the redshift observed for a particular set of lines, it is possible to reconstruct the distance of the gas in question: in an expanding universe like ours, the (cosmological) redshift is directly related to an object's distance from us. In this way, ASPECS was able to probe the cosmological volume of the Ultra-Deep Field, mapping gas-cloud positions in three dimensions.

Keeping track of galaxies – and their molecular gas

The estimate can be made more precise by combining it with another method. Because cosmic dust acts as a catalyst in the formation of molecular hydrogen, there is a correlation between the amount of dust and molecular hydrogen present. ALMA can measure the thermal radiation from that dust in parallel to the CO, allowing for a cross-check. 

In the end, the ASPECS data provided the deepest view of the dusty universe to date, and was able to pinpoint which of the many galaxies visible in the Hubble Space Telescope observations are rich in molecular gas and dust: the material that is essential for star formation to proceed. These galaxies showed a wide range of physical properties: many of them are "normal galaxies" (with average stellar masses and star formation rates), but others are classified as starbursts (with unusually high star formation activity) or quiescent galaxies (unusually low activity).

Reconstructing the star-production supply chain

Once they had made their observations, Fabian Walter and his colleagues were ready to reconstruct the history of molecular hydrogen supplies throughout cosmic history – more specifically: from about 2 billion years after the big bang (nearly 12 billion years ago) to the present. To this end, they drew together the data from previous studies, namely data about atomic hydrogen and about the total mass of all stars in a given epoch. They also compared their findings with large-scale simulations of cosmic history from the big bang to the present.

If you are not an astronomer, the resulting history might not sound all that exciting, compared to the human history you know and can relate to. But for astronomers, it captures deep truths about how our cosmos has changed over time. In that history, the amount of molecular hydrogen steadily increased until about 10 billion years ago, about 4 billion years after the big bang (at about cosmic redshift z=1.5, to use the astronomers' preferred way of denoting a cosmic epoch), with the inventory almost doubling within 3 billion years. This evolution had already been suggested by previous studies. But it is only now that the observations were sufficiently accurate for the firm conclusion that cosmic gas density rises and falls over cosmic time. That rise, then, corresponds to the Golden Age of star formation: With plenty of raw material just waiting to be turned into blazing suns, and with half of the stars that ever existed coming into being in that first third of cosmic history. At the high point, there was about as much molecular hydrogen as there was atomic hydrogen.

What is behind the history of star formation?

In comparing their data with simulations, the astronomers found that behind those boom times was a combination of factors. Galaxies are only the visible tip of the iceberg – their backbone, so to speak, are accumulations of dark matter, matter that does not interact with electromagnetic radiation and thus remains invisible to direct observations. Dark matter accounts for about 80% of all mass in the universe. Just like all other matter, dark matter started out distributed almost perfectly homogeneously through the cosmos shortly after the big bang, but has clumped, and thus become increasingly inhomogeneous, owing to mutual gravitational attraction. In the present-day universe, on a scale of hundreds of millions of light-years, dark matter forms a sponge-like network of filaments, sprinkled with particularly dense regions known as halos. 

Galaxies formed as ordinary matter, mostly hydrogen gas, was drawn into those halos, following their gravitational attraction: First, plasma falls onto halos from the huge reservoir in intergalactic space, cooling down to form atoms. This process replenishes the supply of atomic hydrogen within galaxies. Then, the atomic hydrogen is drawn towards the centers of galaxies, cooling down further until it forms molecular hydrogen, and eventually stars. Through the ASPECS observations, Walter and his colleagues were able to quantify these gas flows as a function of cosmic time.

Looking towards the future, as halo growth slows down and less hydrogen plasma is drawn onto galaxies, star production becomes less and less effective. At the present time, galaxies form stars at a mere tenth of the production rate of the Golden Age. Production rates have been in sharp decline for the past 9 billion years. Based on their observations, Walter and his colleagues predict a continuing trend: Over the next 5 billion years, the molecular gas reservoirs will shrink by an additional factor of 2, while the total mass of stars in the universe increases by a mere 10%. In this picture, star production would eventually cease altogether.

Next steps

The ASPECS observations were designed to be very sensitive, by summing up the light from a larger region in each image pixel. But that automatically meant they could not distinguish smaller details – such as mapping the molecular hydrogen within each galaxy. But now that the combination and ASPECS and Ultra-Deep Field images has enabled astronomers to pinpoint its gas-rich and dust-rich galaxies, the next step will be to take a closer look at those galaxies individually. ALMA has a high-resolution mode that is ideal for that kind of close scrutiny.

This would allow Walter and his colleagues to compare the structure of the molecular gas and dust in those galaxies to the distribution of stars – are the two directly related? Do we indeed find molecular gas and dust in the same region where we find young stars? The more detailed measurements would also yield information about key parameters such as the kinematics, temperature and density of the gas.

With that new ALMA data, plus complementary results from observing campaigns of the Ultra-Deep Field planned for the upcoming James Webb Space Telescope (JWST), the astronomers hope to reconstruct the cosmic history of star formation in even more detail. 

Background information

The results described here have been accepted for publication as F. Walter et al., " The Evolution of the Baryons Associated with Galaxies Averaged over Cosmic Time and Space" in The Astrophysical Journal. 

Original article for this press release:

The ASPECS collaboration is presenting their results on a new website, that will be open to the public from 24 September 2020 onwards. The website also features images, videos and an interactive presentation of the ASPECS results: 

The research was carried out by MPIA's Fabian Walter, Marcel Neeleman and Hans-Walter Rix in collaboration with Manuel Aravena (Universidad Diego Portales, Chile), Chris Carilli (NRAO, Socorro, USA) and Roberto Decarli (INAF, Bologna , Italy). 

The research is part of the of the project Cosmic_Gas that has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (Grant agreement No. 740246). 

The Atacama Large Millimeter/submillimeter Array (ALMA), an international astronomy facility, is a partnership of ESO, the U.S. National Science Foundation (NSF) and the National Institutes of Natural Sciences (NINS) of Japan in cooperation with the Republic of Chile. ALMA is funded by ESO on behalf of its Member States, by NSF in cooperation with the National Research Council of Canada (NRC) and the National Science Council of Taiwan (NSC) and by NINS in cooperation with the Academia Sinica (AS) in Taiwan and the Korea Astronomy and Space Science Institute (KASI). ALMA construction and operations are led by ESO on behalf of its Member States; by the National Radio Astronomy Observatory (NRAO), managed by Associated Universities, Inc. (AUI), on behalf of North America; and by the National Astronomical Observatory of Japan (NAOJ) on behalf of East Asia. The Joint ALMA Observatory (JAO) provides the unified leadership and management of the construction, commissioning and operation of ALMA. 

Other Interesting Articles

Go to Editor View