Loosening the Hubble Tension

For almost a decade, astronomers have been struggling with a nagging mismatch between two different ways of determining the Hubble constant — a measure of the current expansion rate of the universe. This mismatch, known as the Hubble tension, has led to claims that new physics might be needed to solve the issue. (Read about the “constant controversy” in the June 2019 issue of Sky & Telescope.)

But a detailed analysis of a new set of James Webb Space Telescope (JWST) observations now suggests that the problem may not exist. “As Carl Sagan said, extraordinary claims require extraordinary evidence,” says Wendy Freedman (University of Chicago), “and I don’t see extraordinary evidence.”

The Hubble constant (H₀) is expressed in units of kilometers per second per megaparsec (km/s/Mpc), where 1 megaparsec equals 3.26 million light-years. It’s the ratio between a galaxy’s apparent recession velocity (resulting from cosmic expansion) and its distance.

The traditional, so-called “local” way of determining the Hubble constant is by gauging precise distances to galaxies using standard candles, for which the apparent brightness can be compared to the known true luminosity. That distance can then be compared to the galaxies’ redshifts – a proxy for the amount of cosmic expansion that occurred as the light traveled to Earth. Over the past decade or so, a team led by Adam Riess (Johns Hopkins University) has used this method to arrive at a value of H₀ of approximately 73 km/s/Mpc, mainly based on observations with the Hubble Space Telescope.

Another way is to analyze the statistical properties of patterns in the cosmic microwave background (CMB), also known as the Big Bang’s afterglow. Using cosmological models, astronomers can use temperature fluctuations in the CMB to calculate the current expansion rate of the universe. From the European Planck mission’s data, this method yields a Hubble constant of 67.4 km/s/Mpc.

The tension between these two methods has seemed real and unbridgeable, leading to speculations about possible flaws in our fundamental understanding of the universe. According to cosmologist Richard Ellis (University College London), “it would be astonishing if the local value [for the Hubble constant] departed from the CMB measures as it might imply some new, hitherto undiscovered, physics.”

But in a recent paper posted to the arXiv astronomy preprint server, a team led by Freedman presents new JWST data on 11 galaxies that “do not lend strong support to the suggestion that there is missing fundamental physics in the early universe,” as the authors write. (The paper has been submitted to the Astrophysical Journal, but has not yet passed peer review.) Their comprehensive analysis points to a value of H₀ between 68.4 and 71.5 km/s/Mpc, which they say is consistent with the current standard model of cosmology.

Riess and his colleagues used Cepheid variable stars as their standard candle of choice to measure galaxy distances, since these bright and relatively young supergiant stars have periods of variability related to their absolute luminosity. Freedman used these stars, too.

However, Freedman’s team also looked at two other types of standard candles. One is a class of old, low-mass stars that undergo a sudden “flash” of helium fusion in their cores upon reaching the end of their red-giant evolutionary phase; these stars are part of the so-called tip of the red giant branch (TRGB). Another is a particular type of carbon-rich pulsating stars, known as JAGB stars (short for J-region asymptotic giant branch).

According to Freedman, these two methods are more accurate than using Cepheids, although she admits that there is no single perfect method. “Cepheids are not simple standard candles,” she says. “They’re complex.” In analyzing the results, astronomers need to correct for their temperatures and for their abundance of heavy elements.

Moreover, since Cepheids are relatively young, they are mainly found in the densely populated, dusty spiral arms and inner disks of galaxies, so you need to understand the role of dust absorption and of crowding, where a Cepheid may appear brighter than it really is because its light is blended with the light of other, nearby stars, depending on the angular resolution of the telescope.

In contrast, red-giant and JAGB stars, although intrinsically fainter than Cepheids, are found in the outer disks and halos of galaxies. Using these stars yields galaxy distances that are in “superb accord” with each other, says Freedman, while the Cepheid method arrives at somewhat smaller distances and a corresponding higher value for the Hubble constant.

JWST’s angular resolution is four times higher than Hubble’s, and a recent JWST study by Riess’s team indicates that crowding effects do not play a decisive role.  Nevertheless, Freedman thinks that the divergent results for Cepheids as found by Riess and his colleagues could potentially result from the many associated complexities with this kind of star. “Every method is plagued by systematic errors,” she says, “some of which may be unknown.”

Ellis agrees and is particularly impressed by Freedman’s TRGB work. “JWST is the new kid on the block,” he says. “I really find [her] case convincing. It definitely raises the question of whether the error bar on the Cepheid-based value for H₀ has been underestimated due to contamination by crowding.” That would bring the various distance estimates in closer agreement with each other, loosening the Hubble tension.

The value for the Hubble constant presented by Freedman’s team is still a bit higher than the CMB-based value, but the error bars now overlap. “The CMB people are convinced that everything [in their analysis] is understood,” says Freedman, “but it’s very challenging to arrive at one-percent precision.” Moreover, to turn local galaxy distances into a value for H₀ also involves the precise calibration of Type Ia supernovae as standard candles for much more remote galaxies, which might introduce yet other uncertainties. “This is all really hard work.”

The hard work will continue for years to come. “So far, we’ve only scratched the surface,” says Freedman. “By observing more distant galaxies with JWST, we’ll get to the bottom of it.”

Within the next decade, large facilities like the European Extremely Large Telescope and NASA’s Roman space telescope will also be able to measure galaxy distances in much larger volumes of space, while the Vera Rubin Observatory is expected to discover some 300,000 new distant supernovae per year.

Says Freedman: “The path is clear; we’re only just beginning.”

---

Editorial note (August 23, 2024): Since this story was published, Adam Riess's team has posted a rebuttal on the arXiv repository, making the claim that the TRGB and JAGB methods result in a Hubble constant consistent with the one measured using Cepheids. Read their results here: https://www.arxiv.org/abs/2408.11770.