THEY were, on the face of it, inexplicably massive. In February, astronomers announced that among the extremely distant galaxies spotted by the James Webb Space Telescope (JWST), six appeared much brighter, and therefore much larger and more mature, than anyone had expected. One galaxy, pictured as it was just 700 million years after the big bang, contained more than 100 billion stars – roughly the same number that our galaxy, the Milky Way, has amassed over 13 billion years.
Once again, cosmologists were confronted with observations that are impossible to explain according to our best model of how the universe evolved. Not enough time had elapsed to have brought together that amount of matter and turned it into this many stars. At the time, at the University of Texas in Austin argued that the scenario posed a “serious challenge” to our understanding of the cosmos.
And yet, as astronomers pored over the data, it became clear that the current cosmological model is resilient, as it has proven so many times before. Or is it? Because although some analyses indicate that these six galaxies aren’t as massive as first thought, others suggest that they might be even bigger. This indicates that, depending on follow-up observations, we may yet have to remake cosmology – most likely by throwing new cosmic ingredients into the mix to explain the apparent paradox.
Advertisement
“It basically means you’re seeing galaxies before they have time to assemble,” says , an astrophysicist at the Cosmic Dawn Centre in Denmark. “If this is really true, this does mean the standard model of cosmology is broken.”
When physicists talk about a cosmological model, they are referring to a set of equations that describe the evolution of the universe in response to the matter and energy it contains. To build the prevailing story of how the cosmos we see today came to be, known as the standard model of cosmology, they started with Albert Einstein’s general theory of relativity, which casts gravity as the result of mass warping space-time.
Those equations tell us that, although the universe expands overall, specific regions of space can become dense enough to pull matter together through gravity, forming galaxies. Cosmologists fed the equations into supercomputers, along with a list of “ingredients” that reflect the composition of the universe, and ran the simulations. By comparing the galaxies that pop out with our observations, they were able to tweak the model, over the course of decades, to better resemble what the universe actually looks like.
The standard model of cosmology
What we have ended up with is a model in which the universe was sculpted through a combination of gravity, familiar matter and two exotic ingredients. These exotic components are dark matter, which is required to provide a gravitational pull beyond that which known matter can muster, and dark energy, thought to be powering the accelerating expansion of the universe.
Both remain hypothetical, in the sense that we have yet to detect or otherwise identify either. But astronomers nevertheless reckon they have a handle on their characteristics because of what they do. In the case of dark matter, they believe it is composed of massive, sluggish particles that together outweigh normal matter by around 5:1. This is known as cold dark matter, or CDM. Dark energy, for its part, is assumed to be an unchanging energy field – an idea that Einstein toyed with in his equations under the guise of a parameter called lambda. Hence, the standard model of cosmology is known as lambda-CDM.
To be clear, lambda-CDM is remarkably successful. It does an excellent job of explaining the growth of galaxy clusters and other large-scale structures in the universe. But given the mysteries surrounding dark matter and dark energy, researchers are always on the lookout for fresh observations that would help them pin down the characteristics of those ingredients to improve their model.
That is why they were excited to see JWST’s observations of young galaxies. But when at Swinburne University of Technology in Australia and his colleagues used tried-and-tested calculations to glean the masses of these objects based on their overall luminosity, they got more than they bargained for. They found the galaxies had grown so massive so quickly that they sit right on the edge of mathematical possibility in a lambda-CDM universe.
When Boylan-Kolchin got wind of Labbé’s results, he immediately ran what he calls . This involved looking at how much matter it was possible to accumulate in a dark matter halo in the early universe – haloes being large clumps of dark matter that are thought to corral ordinary matter, usually in the form of gas, to create galaxies. He discovered that it was just possible to pull together something with the mass of the Milky Way.
However, to reproduce Labbé’s observations, the galaxy would have had to convert essentially all of its atomic matter into stars. And that is a big ask, to put it mildly. “The galaxy would have to be forming stars even in the far outreaches of these collections of dark matter, where the gas is pretty diffuse and just starting to trickle in,” says Boylan-Kolchin.
While astronomers expect a lot of stars to form in a galaxy’s central region, the outskirts are usually too diffuse to ignite much activity at all. This drastically drags down the efficiency at which a galaxy converts its gas into stars. Typically, star formation in a large galaxy involves just 10 per cent of this gas. In the case of the JWST galaxies, Boylan-Kolchin found that these would have had to be running at 100 per cent star formation efficiency, converting all the gas. “That’s very unrealistic,” he says. “It is basically impossible.”
The upshot is that the galaxies themselves appear to be impossible in the context of the universe as we thought we knew it. But their appearance wouldn’t be the first observation that threatens to break the standard cosmological model. A discrepancy between the expansion rate of the universe as calculated in the relatively nearby universe versus what cosmologists see in the distant reaches of space has been simmering for years now. It is known as the Hubble tension, and if it turns out to be real, cosmologists will almost certainly have to drastically modify the lambda-CDM model to accommodate some sort of early burst of dark energy.
The anomalous young galaxies seemed to be pulling in the same direction. An early burst of dark energy would mean there had to be more dark matter and more ordinary matter in the universe than we thought. More dark matter means larger haloes, and larger haloes mean more efficient star formation.
The question is whether the calculations of the galactic masses stand up to scrutiny. “If these results are really right, there seems to be something seriously wrong [with traditional lambda-CDM],” says Boylan-Kolchin. “So we better confirm or reject these results as quickly as we can.”
The thermodynamics of star formation
Trying to do so has engaged many astronomers, and one line of work has already pointed to a way to ease the discrepancy. To grasp how, first you have to understand that the standard way to estimate a young galaxy’s mass is to look at its total brightness and calculate how many stars would be needed to make it that bright. While this sounds perfectly reasonable, it assumes that you know how various factors influence star formation.
To better establish those factors, Steinhardt has examined how big early galaxies should be expected to grow in light of the nuances of star formation in the early universe – an epoch in which the interplay between gravity and thermodynamics, or the laws of heat and energy, may not have been the same as today.
In star formation, the temperature of interstellar gas is crucial in determining the mass distribution of stars that form in a given cluster – that is, how many high-mass stars you get compared to low-mass stars – which is described by what astronomers call the “initial mass function”.
The reason Steinhardt wanted to examine Labbé’s galaxies is that the initial mass function astronomers universally apply is derived from the conditions in the Milky Way as it is today, whereas we know that star-forming gases would have been considerably hotter in the early universe. That would have inhibited the formation of low-mass stars, changing the initial mass function, the result of which would be a reduction of the mass of any given galaxy as a whole.
Sure enough, when he included such factors in his calculations, Steinhardt found that . “We get the masses to come down for some of Labbé’s galaxies by between a factor of 10 and 100,” he says. The upshot is clear: “You might be able to still make them from the [dark matter] haloes that you can get under lambda-CDM.”
That sounds a lot like another escape for cosmology-as-usual. But lambda-CDM isn’t yet in the clear because Clara Giménez-Arteaga, a PhD student also at the Cosmic Dawn Centre, has performed another analysis of other early galaxies spotted by JWST – not the six Labbé had looked at – and got a very different result.
She was able to take an alternative approach to estimating the mass of young galaxies thanks to JWST’s unprecedented optics, which can resolve even these far-off star clusters into collections of pixels – rather than single pixels containing less detail. That means you can estimate the number of stars and their masses in each pixel, then add them up to compare them with the value from the overall luminosity approach that Labbé used. “Thinking about this beforehand, I thought I should get about the same answer,” says Giménez-Arteaga. What she found, however, was surprising: .
The power of the James Webb Space Telescope
The reason is that, in effect, treating each galaxy as a single unresolved pixel, rather than as a collection of pixels, hides the fact that star formation may not be taking place uniformly across the galaxy. The unresolved method causes brighter, newer stars to outshine dimmer, older stars, masking them from view and lowering the estimated mass, says Giménez-Arteaga. By treating each individual pixel as its own region, astronomers can see the vast number of long-lived, low-mass stars that have been created in previous rounds of star formation. “The physical mechanism that makes this happen can only increase the resolved mass,” she says.
Giménez-Arteaga is yet to apply this method to the galaxies Labbé analysed, but given that the technique has been shown to increase the calculated mass of similar early galaxies, it would almost certainly do the same for the six in question – and would therefore intensify the contradiction with lambda-CDM.
Clearly, cosmologists need to get the true measure of these suspicious galaxies. The fate of the universe, or at least our understanding of it, depends on the answer. The good news is that, unlike the Hubble tension, a case that seems set to rumble on, we can expect definitive answers in the not-too-distant future.
Almost all the analyses done so far rely on images of the galaxies, which require astronomers to estimate a number of quantities such as age, distance and mass. But to accurately pin these down, you need spectra – where the collected light from an object is split into its constituent wavelengths for more detailed analysis. This is the next step in the process and, fortunately, is precisely what JWST was made for.
Unlike its predecessor, the Hubble Space Telescope, JWST is designed to capture light from the really distant universe, which has been dramatically stretched into the infrared region by the expansion of the universe. “JWST offers, for the first time, good quality spectroscopy covering the crucial wavelength range,” says at the University of Oxford, who is a member of JWST’s near-infrared spectroscopy instrument team.
With infrared spectroscopy, we can determine accurate distances and ages for the galaxies. Assuming that each galaxy is confirmed to be at the distance currently estimated – as most astronomers seem to confidently expect – spectroscopy will also allow us to test Steinhardt’s ideas by investigating the temperature of the interstellar medium at each galaxy.
As a proof of concept, Bunker and his collaborators recently released that had showed as a faint red dot in earlier Hubble data and was earmarked for further investigation. The results exceeded anything he had dreamed of. “We never thought we’d get such a beautiful spectrum,” he says.
The upshot is that the galaxy, which is only about 700 million years old, appears to have experienced a short yet intense burst of star formation, followed by a rapid slowdown about 10 to 20 million years before the time of observation. Particularly interesting, says Bunker, is that the mass they calculated came in at around 200 times smaller than the Labbé sample.
Bunker says this isn’t a direct refutation of the idea that those six galaxies could break cosmology, because the galaxies he analysed are in a different part of the sky. “It’s possible lambda-CDM is broken, but the jury is out until we have the spectroscopy,” he says.
So, the plot continues to thicken. For now, lambda-CDM has a stay of execution. But even if it survives this current crisis, it will face the chop at some point, says Steinhardt. “Lambda-CDM is a placeholder,” he adds, meaning that until we understand the true nature of dark matter and dark energy, we are simply using the most generic examples of both in the model. In that context, it is perhaps surprising that such a simple model has been able to explain the entire universe for so long.
And yet it seems clear, at this stage, that JWST’s ability to peer into the furthest reaches of the cosmos means it is going to keep finding things that put lambda-CDM under pressure. “It has already been a game changer,” says Bunker. “We’re now routinely getting high-quality spectra for which we can infer the properties of galaxies just a few hundred million years after the big bang.”
Stuart Clark is an astronomy journalist based in London. His latest book is Beneath the Night
Â鶹´«Ă˝ Discovery Tours
Unpick the mysteries of the universe at Jodrell Bank Observatory, UK
