Have we been measuring the expansion of the universe wrong all along?

WENDY FREEDMAN is staring down the universe. For 40 years, she has been digging into the biggest secrets of the cosmos, patiently whittling down uncertainties to find the value of a number that defines the expansion of the universe, determines its age and seals its ultimate fate.

Freedman, who works at the University of Chicago, studies the Hubble constant, a number that represents how fast the expansion of the universe is accelerating. We have known about this escalating expansion since 1929, when US astronomer Edwin Hubble found that the more distant an object was, the faster it seemed to be moving away from us.

That is when things got tricky. Pinning down the numbers requires accurate measurements of astronomical distances. In Hubble’s era, astronomical images were taken by shining light through a telescope onto a photographic plate. Calculating distances from those images was difficult and imprecise.

In the 1980s, as Freedman was finishing her PhD, digital photography was getting ready to revolutionise astronomy as a whole, and measurements of the Hubble constant in particular. “That’s really what spurred me,” says Freedman. In the decades since, her work has been key to the development of the Hubble tension – the perplexing way that the two main ways of measuring the Hubble constant give us different values.

Now, after Freedman has spent decades focusing on this problem, something curious is happening. Her newest results suggest there may be no problem after all. If this is the case, it will render pointless decades of work exploring new physics that could explain the discrepancy. Luckily, Freedman isn’t afraid of a little controversy.

The Hubble constant is a big piece of the cosmological jigsaw that, when put together, tells us about the history and future of the universe. If we know how quickly the expansion of the universe is accelerating, that hints how big the cosmos is, how old it is and how it began. Looking ahead, the Hubble constant determines whether the universe will expand forever or collapse in a big crunch.

A cosmic ladder

Before Freedman came on the scene, there were two main estimates of the Hubble constant. French astronomer Gérard de Vaucouleurs found it to be 100 kilometres per second per megaparsec – a megaparsec being equal to 3.26 million light years. But US astronomer Allan Sandage found it to be much lower, at about 50 km/sec/mpc. The two were locked in a fierce debate that had been raging for decades, until the 1980s. At that time, Freedman was a postdoctoral fellow at the Carnegie Observatories in California, where Sandage was a professor. Although well-respected, Sandage was sometimes feared for his anger. He would “go non-linear” when he was contradicted, says Princeton University cosmologist and Nobel laureate Jim Peebles.

When Freedman’s first results started coming in from her new observations, they indicated a Hubble constant closer to 80, contradicting two of the most renowned cosmologists in the world. Sandage wasn’t thrilled, to say the least. “It’s hard being contradicted by a young upstart – what do they know?” says Peebles. “Well, Wendy knew a lot. She is an absolute model of toughness.”

You might think it would be a scary place to be, going against the most prominent astronomers of the time. “I didn’t find it scary,” says Freedman. “It was fun.” She knew her results were clear. Her calculations used observations of Cepheids, stars that pulsate regularly. Sandage and de Vaucouleurs used these stars, too. But crucially, Freedman’s early observations were among the first to be corrected to factor in the dust between us and the Cepheids, making the calculated distances more accurate than those in previous work.

Northern winter constellations and a long arc of the Milky Way are setting in this night skyscape looking toward the Pacific Ocean from Point Reyes on planet Earth's California coast. Sirius, alpha star of Canis Major, is prominent below the starry arc toward the left. Orion's yellowish Betelgeuse, Aldebaran in Taurus, and the blue tinted Pleiades star cluster also find themselves between Milky Way and northwestern horizon near the center of the scene. The nebulae visible in the series of exposures used to construct this panoramic view were captured in early March, but are just too faint to be seen with the unaided eye. On that northern night their expansive glow includes the reddish semi-circle of Barnard's Loop in Orion and NGC 1499 above and right of the Pleiades, also known as the California Nebula.

Cepheids have a pulsation period that is directly related to their absolute luminosity, so comparing that with how bright they appear from Earth can tell us how distant they really are. Those measurements can be used to extrapolate outwards, using supernovae in the same galaxies as the Cepheids. This method is known as the cosmic distance ladder, because each step along the way builds up to the next.

Over the decades since her initial results, Freedman’s measurements have held up. Cepheids have remained the main tool by which we measure the expansion of the universe in the area relatively close to Earth – one of the first steps on the distance ladder. In fact, observing them, and measuring the Hubble constant to 10 per cent accuracy, was one of four so-called key projects of the Hubble Space Telescope, launched in 1990. “The director of the project asked, if Hubble fell into the ocean a month after it started observing, what were the projects we’d really want done,” says Freedman. “Every telescope has a big project that’s its goal, and settling this debate was the big problem that was sort of before us.”

The project was a resounding success. Using the telescope, Freedman’s team measured a Hubble constant of about 72 km/sec/mpc – right in between the earlier estimates. Even as instrumentation has improved since the key project paper was published in 2001, the value we get from measuring Cepheids has remained similar, with the most recent observations putting it around 73 km/sec/mpc.

But that isn’t the end of the story, because Cepheids aren’t the only way to measure the Hubble constant. Another problem came along when astrophysicists started observing the cosmic microwave background (CMB), relic light left over from the moments after the big bang. By observing this light and extrapolating forwards in time based on our best models of the universe, we can predict what the Hubble constant ought to be today.

“The CMB is really predicting the Hubble constant starting at the other end of the universe, using our story of cosmology and physics,” says Adam Riess at Johns Hopkins University in Maryland, who won a Nobel prize for his work on the expansion of the universe. And these measurements from the other end of the universe, which are extraordinarily precise, don’t agree with the measurements from this end. They put the Hubble constant at around 67 km/sec/mpc.

While the CMB measurements themselves are extremely precise, the value of the Hubble constant we get from them is calculated using physicists’ standard model of the cosmos – a set of equations that fit just about everything, from general relativity to the effects of dark matter. “I continue to be astounded at how well theory and observation fit,” says Peebles. “But eventually, we’re going to find something that doesn’t fit, and maybe this is it.”

Tension or no tension?

If both the Cepheid measurements and the CMB measurements are correct, a problem known as the Hubble tension, then something is wrong with our understanding of the cosmos. And if we shift one part of that understanding – for instance, the way the universe inflated after the big bang – it won’t just solve the Hubble tension. It will have a knock-on effect on other factors, in ways we don’t know how to account for. “If anything changes, it’s going to change everything,” says Freedman.

The big question now is whether it is time to change everything. Riess claims that it is. He says the CMB measurements are solid. His group has made many Cepheid measurements, and the tension between the two methods remains. “It’s hard to ignore that basically all the precise measurements are coming in higher than the CMB,” he says.

Other cosmologists aren’t so convinced, not least because of how difficult it is to explain how the two values could be different. “People are trying to come up with ways to explain this, and almost 1000 papers later, they haven’t,” says Freedman. “It’s much more interesting to say there’s new physics than there are systematic uncertainties, but that doesn’t mean it’s how the universe is.”

Nailing down whether the Hubble tension is real needs more measurement methods, says Freedman. Relying solely on Cepheids, it is impossible to know if we are misunderstanding those stars in some basic way that throws measurements off. The uncertainties in the CMB measurement are below 1 per cent. Ideally, uncertainties in measurements of the local Hubble constant would be similarly low. “Right now, there’s disagreement about the errors of the errors,” says Freedman – it isn’t clear where exactly the uncertainties are now, but she doesn’t believe we are at 1 per cent yet. “I think to be certain that the errors are less than 1 per cent, you need more than one type of measurement, more than one instrument, more than one technique,” she says. “I just think we need to be a little bit patient.”

That is why Freedman turned to a different source, called tip of the red giant branch stars. These are the brightest stars in a group called the red giants, which make up a branch on the Hertzsprung-Russell diagram, a plot of stars’ temperature against their luminosity. Tip of the red giant branch stars are simpler than Cepheids, and we have a better understanding of the physics that determines how bright they are and their colours. They are also extremely common and located throughout galaxies, whereas Cepheids are generally more concentrated towards the centres. This means we don’t have to worry too much about other stars or dust contaminating our images, as we can simply look at stars that are in less busy areas.

These stars have been used to measure distances for a long time, but they fell out of favour because they are dimmer than Cepheids. Now that telescopes have improved, the simpler stars are coming back to the fore. “This method has always been there, sort of looming in the background,” says Barry Madore at the Carnegie Institution for Science in California, Freedman’s husband and frequent collaborator. “Its simplicity is just crushingly, mind-blowingly good. There are so few things that can go wrong with it.”

So when Freedman’s most recent measurements using the tip of the red giant method yielded a Hubble constant of 69.8 – right between the CMB and the Cepheid numbers – it sowed fresh seeds of doubt among the growing certainty in the community that believe in the Hubble tension. “It was quite a surprise that she came up with these results that are in between, and sort of mitigated the growing feeling that we had that there really was an issue,” says Brent Tully at the Institute for Astronomy in Hawaii. “I think that Wendy makes a good case that there are still outstanding problems.”

Once again, Freedman finds herself at the centre of the debate. “We imagined that when we got our result that it would land on one side or the other, but it just didn’t,” says Freedman. “It’s saying this isn’t really tied up yet.” Not everyone agrees – Riess in particular still claims that it actually is tied up, and the Hubble tension is real – but Freedman remains cautious.

And once again, the solution may come from a huge new space telescope. This time, it is the James Webb Space Telescope (JWST), which launched at the end of 2021 and is expected to begin observations around July this year.

The JWST has a larger mirror than Hubble, so it can look at more distant objects, and unlike Hubble, the JWST is an infrared telescope. That isn’t ideal for Cepheids, which are relatively blue in colour. It will help us improve the precision of our current Cepheid observations, but it is unlikely we will find any more distant ones. The tip of the red giant branch is a different story. As their name suggests, these stars are red, so when viewed in the infrared, they appear brighter than the surrounding stars. “I think that we’re really gonna nail this thing through the tip of the red giant branch,” says Tully.

Freedman has been approved to observe those stars with the JWST once it starts up, and in the meantime she is observing them and Cepheids from the ground, as well as looking into a potential new method using stars that are extremely rich in carbon. “If you knew the answer, you’d stop,” says Madore. “We don’t know the answer.”

The solution to the Hubble tension may come from measurements, but it will be deeply entangled in theoretical calculations about the cosmos as well. “You can’t just measure your way to an answer – these things go hand in hand,” says Riess. If the tension isn’t real, we need to figure out what we misunderstood in the first place. If it is real, there is a far deeper well of misunderstandings. Should we learn that our standard model of the universe is incomplete, we will doubtless find other places where that model breaks down.

“If the Hubble anomaly is real, then there have to be other anomalies,” says Peebles. To build a better model of the cosmos, we will have to hunt each anomaly down, prove it is real and figure out where it came from. All this means more hard work, and more waiting. But Freedman doesn’t mind – it’s business as usual for her. “The universe is doing whatever it’s doing,” she says. “It doesn’t care when, or whether, we answer our questions about it.”

Source: Link