How the Webb and Gaia missions bring a new perspective on galaxy formation

In a feat of galactic archeology, astronomers are using ever more detailed information to trace the origin of our galaxy—and to learn about how other galaxies formed in the early stages of the Universe. Using powerful space telescopes like Gaia and James Webb, astronomers are able to peer back in time and look at some of the oldest stars and galaxies. Between Gaia’s data on the position and movements of stars within our Milky Way and Webb’s observations of early galaxies that formed when the Universe was still young, astronomers are learning how galaxies come together and have made surprising discoveries that suggest the early Universe was busier and brighter than anyone previously imagined.

The Milky Way’s earliest pieces

In a recent paper, researchers using the Gaia space telescope identified two streams of stars, named Shakti and Shiva, each of which contains a total mass of around 10 million Suns and which are thought to have merged into the Milky Way around 12 billion years ago.

These streams were present even before the Milky Way had features like a disk or its spiral arms, and researchers think they could be some of the earliest building blocks of the galaxy as it developed.

"What's very interesting is that we are able to detect these structures at all from such ancient times,” said lead researcher Khyati Malhan of the Max Planck Institute for Astronomy (MPIA). “These very massive fragments came in and they collapsed under their own gravitational force, and they basically formed the proto Milky Way galaxy.”

This happened when the Universe was still young, with the earliest galaxies forming only around 13 billion years ago. When these groups of stars were coming together to form what would be the Milky Way, it's debatable whether the group they joined could even have been called a galaxy. While there are broad gravitational requirements for a certain mass of stars to hold itself together, there’s no precise definition of when a group of stars can truly be called the start of a galaxy.

“When is a city a city?” said coauthor Hans-Walter Rix, also of MPIA. “This is why there isn’t an epoch when the galaxy formed. It’s been a continuous process.”

The Milky Way as a test case

With so much still to learn about the formation of galaxies, it makes sense to start with our own Milky Way galaxy as a test case. The Milky Way is “a tremendously average galaxy,” Rix said. When compared to the rest of the Universe, “Half the stars live in bigger galaxies, half the stars live in smaller galaxies.”

What makes the Milky Way useful is that we have unique access to it, providing the ability to see individual stars within it. That means researchers can identify large groups of stars that seem to have originated together with similar ages and levels of heavier elements. Looking at each of these groups lets them track how the galaxy was pieced together.

There are two main ways that stars enter galaxies. In the first, large clouds of diffuse gas are present within an existing galaxy, and this gas condenses so that stars form within it. Alternately, stars that form in a satellite galaxy can be dragged into the main galaxy.

Today, we most often see the formation of stars within gas clouds, with around 90 percent of the stars we see today formed in this way. But at earlier stages of the Universe, the satellite accretion option was much more important, as most stars from this period are thought to have formed in clumps that were then dragged into the young Milky Way.

To understand the Milky Way’s history, astronomers need to trace the origin of these groups of stars and figure out what drew them into the galaxy we know today. “One of the big goals is 'can we reconstruct the early accretion events of these pieces coming together?'” said Rix.

Using Gaia data, the researchers were able to pick out groups of stars with similar orbits that were located toward the center of the galaxy. They are located about halfway between Earth and the galactic center and are found in a thick-wall torus shape revolving around the galaxy's center.

The researchers suspect that the two streams of stars they discovered were some of the final pieces of the Milky Way that were absorbed during the satellite accretion stage, after which star formation within the galaxy took over as the primary driver of stars joining the galaxy. “It looks like Shakti and Shiva are perhaps the last hurrah of that early phase, when it was mostly bits and pieces coming together,” Rix said.

Turning back the clock

It is no easy job to trace the clumps of stars that entered the Milky Way billions of years ago. To understand how a galaxy formed, Malhan said, you would ideally want three pieces of information: the orbits of the stars, their ages, and their chemical compositions. For the Milky Way, surveys like Gaia can provide information on the orbits and compositions of stars, but determining stellar ages remains difficult.

So to figure out where stars originate, researchers can look to the stars’ orbits. Once stars join a galaxy, their locations spread, but the ways in which they move tend to be similar. Rather than looking at a particular area of the galaxy and seeing what stars are within it, the researchers plot out the orbits of stars on a grand scale and identify those that share energies and movements.

Scientists have used the data from Gaia to track the location and motion of stars in our galaxy.

Credit: ESA/Gaia/DPAC

They can also look for telltale signs of older stars by studying their composition. In the very earliest stages of the Universe, the vast majority of all matter was hydrogen and helium. The heavier elements like metals were produced within the cores of the first stars as they underwent fusion; when these stars exploded in supernovae, they spread those metals out and into the Universe.

So, a star that is low in heavy elements (in astronomical terms, it has low metallicity) is an old one. As the Universe evolved, more stars became “polluted” by metals. This is how a star’s composition can indicate its age and why low-metallicity stars are of such interest to those studying the early stages of the galaxy.

“So galaxy evolution is also the history of galactic self-pollution,” Rix said. “And at some level, one can sort stars according to age by the ones that are less polluted or chemically more pristine, which we call metal-poor. They are the old ones. And Gaia has allowed us to actually determine orbits for all these stars.”

Determining exact stellar ages

Looking at a star’s metallicity is a good indicator of its rough age, but to be more precise, researchers search for more detailed information. One indicator that has been useful in studying the Milky Way is looking at other aspects of their composition, such as the preponderance of what are called alpha elements, which can indicate how quickly a star formed. (An alpha particle is another term for the helium nucleus, which fuses to form these elements.)

Due to the nature of the fusion process, long-lived stars produce more iron, while short-lived stars produce more calcium and magnesium, called alpha elements. So there is rapid self-pollution of alpha elements but a more gradual self-pollution of iron. If a high alpha star is identified, that means it formed rapidly. The Shakti and Shiva streams are made up almost entirely of high-alpha, low-metallicity stars. That helps group them as both being very old and having formed around the same time.

Even with this data, however, the measurements of stellar ages are still imprecise. “One is still stuck with using these element ratios as chemical clocks to make qualitative arguments about the age sequence,” Rix said. “Ideally, we'd like to just measure the ages of these stars. There’s the prospect with the next round of Gaia to do this, but we can’t do this yet.”

With the next release of data from Gaia in 2026, there will be a wealth of new information on the ages of stars within the Milky Way. There are stars that have a known relationship between their age and their luminosity, so if this luminosity can be precisely measured, their age can be calculated. These stars are faint and hard to detect, but they'll be included in the next batch of Gaia data. This will enable exciting new areas of research, but it will still be constrained by uncertainties regarding ages.

“We are now getting better at determining the ages of stars to within 10 percent,” Rix said. “But that still means that 11 billion years old and 13 billion years old are barely distinguishable.”

That’s a not-insignificant problem, and even with newer Gaia data, it will still be impossible to know exactly how old a given star is.

So some astronomers take another approach to understanding the formation of the Milky Way: looking beyond our own galaxy and into the Universe, searching for similar galaxies at an earlier stage of formation.

With the James Webb Space Telescope, astronomers have the opportunity to see these distant galaxies in more detail than ever before, allowing them to be examined as models for what the Milky Way could have looked like millions of years ago.

“We have the possibility to test any scenario we concoct for the Milky Way by asking, then let’s envision what the Milky Way would have looked like at redshift 5, 6, or 7,” Rix said. “Which now we can actually see,” thanks to Webb.

Webb’s view of the early universe

Webb’s observations come with their own challenges, though. Determining the age of a galaxy is not a simple matter, even using a powerful tool, for reasons that have as much to do with definitions as with available data.

“How do you define the age of a galaxy? Is it the first time it formed or the most recent set of star formation? We’re talking about an ensemble of objects, all of which are going to have different ages,” said Micaela Bagley of the University of Texas at Austin, a researcher on Webb’s CEERS (Cosmic Evolution Early Release Science Survey) project. CEERS has identified some of the oldest galaxies ever seen, at redshifts of up to 12, which corresponds to as early as just 500 million years after the Big Bang.

These galaxies observed by Webb are too far away to resolve individual stars, so its instruments collect data on the galaxies as a whole. Using cameras like NIRCam, Webb observes many of these distant galaxies as a cluster of pixels, but using its spectroscopy instruments like NIRSpec, it can obtain more detailed data, like signatures of recent or older star formation.

An image stitched together from multiple images captured by the James Webb Space Telescope in near-infrared light.

Credit: NASA, ESA, CSA, STScI

The very distant galaxies observed so far, which are part of the early Universe, are typically showing very active, very recent star formation. Webb can’t track the age of individual stars or groups of stars, but it can look at a galaxy’s history of star formation as a whole—whether a galaxy has had multiple periods of star formation or whether its star formation spiked and then declined, for example.

With these early galaxies, “We’re looking at their populations of stars to try to figure out the history of their star formation,” Bagley explained. “Not necessarily how old it is but how it has gone about its life.”

Early evidence suggests that galaxies have typically experienced star formation in waves, peaking and troughing over time. But all that active star formation at such an early stage in the Universe poses some problems.

A brighter early universe

Early results from Webb have been exciting for many reasons, but astronomers are particularly intrigued to learn that the early Universe is not quite as we imagined it to be. From almost the first week that Webb began its science operations, it has been turning up brighter and more massive galaxies much earlier than anyone expected, and subsequent observations have backed these findings up.

When it comes to galaxies in the early Universe, “Across the board, everything is more,” Bagley said. Webb has found early galaxies that are brighter and more numerous than predicted, and these galaxies have developed in unexpected ways: “There are more black holes than we thought, and there are more dusty galaxies than we thought. These galaxies are able to form dust, which should take a long time, but they’re doing it earlier and more efficiently than we thought.”

The early Universe “is just a much busier place than we thought,” Bagley summed up.

As part of this busy development, galaxies have been found with structures similar to the Milky Way much earlier than had been expected as well. The very earliest galaxies observed by Webb, at around 12 or 13 billion years old, tend to be diffuse and chaotic, with not much structure. But structured galaxies with features like disks, spiral arms, or bars have been found as far back as around 11 billion years (z≈3), far earlier than predicted.

“We thought that it would take a very long time to form a galaxy that was gravitationally stable enough to have a disk and spiral arms like the Milky Way does. But we’re starting to see some of those pretty early in the Universe. Not as early as the first galaxies, but way earlier than we expected,” Bagley said.

Even galaxies from just a few hundred million years after the Big Bang (redshifts of 7 and above), which had previously been only observed by Hubble as blobs of light, seem to be being resolved by Webb as having some kind of structure.

“Galaxies seem to be forming very quickly, then reaching that equilibrium that allows them to rotate and flatten into a disk. And once you’re in that shape, you can start to have structures form, like spiral arms,” Bagley explained.

Simulating the universe

As exciting as these discoveries are, they are proving hard to reconcile with models of galaxy formation we have now. Researchers typically plot out the story of galaxy formation using simulations, aiming to model the presence of features like stars and dark matter and seeing how those interact. Right now, however, none of these simulations exactly match what is seen in the Universe across time.

Given the preponderance of stars and galaxies observed by Webb in the early Universe, the mechanisms of star formation in these simulations need to be made more efficient.

In short, there needs to be some explanation of where all these extra stars are coming from—whether it's because the early Universe was denser than expected and denser galaxies encouraged faster star formation, because the proportion of massive stars to smaller stars being formed is different, or even because the amount of dust present in early galaxies is different, and that makes the stars appear to shine brighter.

It’s not currently known what the best explanation for the prevalence of bright, earlier galaxies is. Though it’s easy enough to tweak a simulation to make it match the possible theories, the problem is that the simulation then over-predicts what is observed today. If a simulation accurately models the efficient star formation seen in the earliest galaxies at redshift 12 or 13, by the time it gets to redshift 7 or so, it is showing more or brighter galaxies than we observe.

“Getting the story correct across the full history of the Universe is really challenging and is probably going to require a lot more observations,” Bagley said. “It’s exciting that we still don’t understand it.”

Georgina is a freelance space writer who covers planetary science, cosmology, and human space exploration. She is based in Berlin, Germany, where she did a PhD in psychology before turning to the world of space. You can see more of her work at www.georginatorbet.com.