Skip to main content

Oldest surviving light reveals the universe's true age

A portion of the Atacama Cosmology Telescope's new image of the cosmic microwave background
A portion of the Atacama Cosmology Telescope's new image of the cosmic microwave background
(Image: © © ACT Collaboration)

 Ancient light from the Big Bang has revealed a precise new estimate for the universe's age: 13.77 billion years, give or take 40 million years.

The new estimate, based on data from an array of telescopes in the Chilean Atacama Desert, also weighs in on one of the most important disagreements in astrophysics: How fast is the universe expanding? Described in two scientific papers, the new result gives a significant boost to one side of the disagreement, though the physicists couldn't prove the other side of the dispute wrong.

Here's the problem: Physicists need to understand the universe's expansion rate to make any sense of cosmology — the science of our whole universe's past, present and future. They know that a mysterious substance called dark energy is causing the universe to expand (at an ever-increasing rate) in all directions.. But when astronomers point their telescopes into space to measure the Hubble constant (H0) — the number that describes how fast the universe is expanding at different distances from us or another point — they come up with numbers that disagree with each other, depending on the method they use.

One method, based on measurements of how fast nearby galaxies are moving away from the Milky Way, produces one H0. Another method, based on studying the oldest light in space, or cosmic microwave background (CMB), produces another H0. This disagreement has left scientists wondering whether there's some important blind spot in their measurements or theories, as Live Science previously reported. These new results seem to show that there weren't any measurement errors on the CMB side.

Related: 9 facts about black holes that will blow your mind

"We find an expansion rate that is right on the estimate by the Planck satellite team," which is another study of the CMB, Cornell University astrophysicist Steve Choi, lead author of one of two new papers, said in a statement. "This gives us more confidence in measurements of the universe's oldest light."

The data from the Planck satellite, released in 2018, were the most important measurements of the CMB before now. With an unprecedented level of precision, they showed how sharply CMB measurements of H0 disagree with measurements based on the movement of nearby galaxies.

These new results recalculated the CMB measurement from scratch using an entirely different set of telescope data and calculations, and came up with very similar results. That doesn't prove that the CMB measurement of H0 is correct — there could still be some problem with the physics theories used to make the calculation — but it does suggest that there aren't any measurement errors on that side of the disagreement.

Related: The 18 biggest unsolved mysteries in physics

Relying on data from the Atacama Cosmology Telescope (ACT) in Chile's Atacama Desert, the researchers tracked faint differences between different parts of the CMB -- which appears to have different energy levels in different parts of the sky. The CMB, which formed as the universe cooled after the Big Bang, is detectable in every direction in space as a microwave glow. It's more than 13 billion light-years in the distance, a relic of a time before stars and galaxies formed. 

By combining  theories on how the CMB formed with precise measurements of its fluctuations, physicists can determine how fast the universe was expanding at that moment in time. That data can then be used to calculate H0.

The ACT methodically scanned half the sky between 2013 and 2016, looking particularly at microwave light. Then researchers spent years cleaning up and analyzing the data with the aid of supercomputers, removing other microwave sources that are not part of the CMB, to stitch together a full map of the CMB. The whole time, they "blinded" themselves to the implications of their work, they wrote in their papers, meaning they didn't look at how their choices affected estimates of H0 until the very end. Only when the full CMB map was complete did the researchers use it to calculate H0.

The new CMB map also offered a new measure for the distance between Earth and the CMB. That distance, combined with a new measurement of how fast the universe has expanded over time, allowed a precise calculation of the age of the universe.

"I didn't have a particular preference for any specific value — it was going to be interesting one way or another," Choi said.

 It's still possible, as Live Science has previously reported, that some error in those theories is messing up the  calculation. But it's not clear what the error would be.

The other approach to calculating H0 relies on pulsing stars known as cepheids, which reside in distant galaxies and pulse regularly. That timed pulsing allows researchers to perform precise calculations of their motion and distances from Earth.

With those direct speed measurements, it's fairly straightforward to come up with a measurement of H0. There are no complicated cosmological theories involved. But it's possible, some scientists have proposed, that our region of the universe is just weirdly empty, and not representative of the whole universe. It's even possible that there are measurement issues with the cepheids, and that these cosmic measuring sticks don't work quite the way physicists expect.

For now, the true H0 remains a mystery. But CMB researchers have more ammunition for their side of the disagreement.

Both new papers describing the new analysis have been published July 14 to the preprint database arXiv and submitted for formal peer review.

Originally published on Live Science.

Join our Space Forums to keep talking space on the latest missions, night sky and more! And if you have a news tip, correction or comment, let us know at: community@space.com.

  • rod
    My observation, the arxiv report is referenced in the space.com article (good). H0 value calculated according to the draft arxiv paper, “Abstract...LCDM is a good fit. The best-fit model has a reduced Chi^2 of 1.07 (PTE = 0.07) with H0 = 67.9 +/- 1.5 km/s/Mpc…”, ref - THE ATACAMA COSMOLOGY TELESCOPE: A MEASUREMENT OF THE COSMIC MICROWAVE BACKGROUND POWER SPECTRA AT 98AND 150GHZ Draft version July 14, 2020.

    My observation, this cosmology calculator (http://www.astro.ucla.edu/~wright/CosmoCalc.html) shows H0 = 67.9, the Hubble time is 14.064E+9 years old. The space.com report shows 13.77E+9, a delta of 2.94E+8 years for the Hubble time age of the universe. I used flat model with cosmology defaults and changed H0 to 67.9. Pinning down the *precise* age of the universe using the Hubble constant and Hubble time, is a model in progress😊
    Reply
  • rod
    FYI folks. It did not take long to have another report, contradict the age of the universe based upon ACT measurements. There is a new report out this morning showing the Hubble constant is 75.1 km/s/Mpc with age of the universe pegged at 12.6E+9 years old now. See, 'New approach refines the Hubble's constant and age of universe', https://phys.org/news/2020-07-approach-refines-hubble-constant-age.html
    "Calculations drawn from observations of NASA's Wilkinson Microwave Anisotropy Probe in 2013 put the age of the universe at 13.77 billion years, which, for the moment, represents the standard model of Big Bang cosmology. The differing Hubble's constant values from the various techniques generally estimate the universe's age at between 12 billion and 14.5 billion years. The new study, based in part on observations made with the Spitzer Space Telescope, adds a new element to how calculations to reach Hubble's constant can be set, by introducing a purely empirical method, using direct observations, to determine the distance to galaxies, Schombert said."

    Cosmology is fun again :)
    Reply
  • rod
    FYI, here is a list of recent reports on the Hubble constant that I have in my home database.

    New approach refines the Hubble's constant and age of universe, https://phys.org/news/2020-07-approach-refines-hubble-constant-age.html, 20-Jul, 75.1 km/s/Mpc

    Oldest surviving light reveals the universe's true age, https://forums.space.com/threads/oldest-surviving-light-reveals-the-universes-true-age.32567/, 20-Jul, 67.9 km/s/Mpc

    NEW 3D MAP OF THE UNIVERSE — AND A GROWING COSMOLOGICAL DEBATE, https://skyandtelescope.org/astronomy-news/new-3d-map-universe-growing-cosmological-debate/, 20-Jul, 69 km/s/Mpc

    Measurement of Hubble Constant: Do Different Cosmological Probes Provide Different Values? https://ui.adsabs.harvard.edu/abs/2020arXiv200611721T/abstract, 20-Jun, 82 km/s/Mpc

    H0LiCOW XIII. A 2.4% measurement of H0 from lensed quasars: 5.3σ tension between early and late-Universe probes, https://ui.adsabs.harvard.edu/abs/2020MNRAS.tmp.1661W/abstract, 20-Jun, 73.3 km/s/Mpc

    New research of oldest light confirms age of the universe, https://phys.org/news/2020-07-oldest-age-universe.html, 20-Jul, 67.6 km/s/Mpc

    New distance measurements bolster challenge to basic model of universe, https://phys.org/news/2020-06-distance-bolster-basic-universe.html, 20-Jun, 74 km/s/Mpc
    Reply
  • Helio
    It's interesting to still see so little attention to some details in favor of others.

    In Aug, 2018, 78% of the IAU vote changed the Hubble Constant to the Hubble-Lemaitre Constant. None of these articles, however, respect the new term.

    Some of the articles, at least one, talks about Hubble in his work to demonstrate expansion. Edwin Hubble never once argued for an expansion, surprisingly. To him that claim was for theorists, not him.

    But, and my very cursory views of the articles, seem to treat the "Hubble Constant" as a constant, but we know it's not. The expansion rate is non-linear. Lemaitre applied Einstein's cosmological constant in his original theory that gave us the Big Bang. Further, he (and Eddington, IIRC) disagreed with Einstein that his cosmological constant should be removed.

    It was in 1998, as we have discussed recently, that an acceleration component is needed for the "constant".

    So, what am I missing? Is the constant a term that has great utility even if it lacks accuracy? Given our margin of errors for distance, etc. aren't fantastic, perhaps so.
    Reply
  • rod
    It seems if H0 is not a constant but perhaps a variable that changes with distance from Earth and perhaps dark energy acceleration, those cosmology calculators, e.g. https://ned.ipac.caltech.edu/help/cosmology_calc.html, may need some tweaking or revision to accurately show the *true age* of the universe :)
    Reply
  • rod
    FYI. The article in this thread is about the *true age* of the universe from CMBR measurements, I posted various report links showing that measuring H0, the Hubble constant is *under review and discussions* and this *constant* impacts the Hubble time calculated for the age of the universe :) One problem I note, all distance measurements used to support BB cosmology and the *true age* of the universe, those distance measurements are not direct like stellar parallax or absolute magnitude compared to apparent magnitude distance calculations. Using a 2 AU baseline (Earth's orbit around the Sun), a star with stellar parallax of 1E-4 arcsecond, could be a bit more than 32600 LY distance from Earth. That means stellar parallax measurements to reach that distance or farther, must be good to 0.1 mas and we do not get negative values but a good, positive stellar parallax measurement for a star. I have not seen this stellar parallax ability (down to 0.1 mas) from Gaia DR2 reports. Thus stellar parallax measurements for various stars is very limited in distance measurement, likely not more than 9,000 LY distance for reliable values. The same holds for red giant stars. A red giant star with apparent magnitude of 26.0 and absolute magnitude of 0, could in theory be seen out to 5.17 million LY distance. Hundreds of millions and billions of light-year distances are indirect and require expanding universe conversions for redshift, converted to Hubble time for the age of the universe in BB cosmology. Even Type 1a supernova distances are limited too (thus they do not show the true age of the universe). https://phys.org/news/2020-07-spectacular-uv-white-dwarfs.html, “Using the Zwicky Transient Facility in California, researchers first spotted the peculiar supernova in December 2019—just a day after it exploded. The event, dubbed SN2019yvq, occurred in a relatively nearby galaxy located 140 million light-years from Earth, very close to tail of the dragon-shaped Draco constellation."
    Reply
  • Helio
    rod said:
    Using a 2 AU baseline (Earth's orbit around the Sun), a star with stellar parallax of 1E-4 arcsecond, could be a bit more than 32600 LY distance from Earth. That means stellar parallax measurements to reach that distance or farther, must be good to 0.1 mas and we do not get negative values but a good, positive stellar parallax measurement for a star.
    We can use angular measurements to get us farther in the case of SN. The mass discharge velocity of 1987A SN, for instance, is easily determined, and it wasn't long before direct imaging could fine tune its angular size. So, the radius of the blast is simply the expansion velocity times time. So with a known radius measurement and a known angular measurement then simple trig was used to confirm the distance of its host galaxy (dwarf) of about 160,000 light years. This was direct imaging. It was also interesting that the neutrino blast arrived only a few seconds before the light, as predicted for light speed and neutrino speeds, IIRC.

    Even Type 1a supernova distances are limited too (thus they do not show the true age of the universe). https://phys.org/news/2020-07-spectacular-uv-white-dwarfs.html, “Using the Zwicky Transient Facility in California, researchers first spotted the peculiar supernova in December 2019—just a day after it exploded. The event, dubbed SN2019yvq, occurred in a relatively nearby galaxy located 140 million light-years from Earth, very close to tail of the dragon-shaped Draco constellation."
    There is a margin in error for all astronomical observations. The greater the distance the greater the margin of error. When the 13.7 Byr. estimate for the universe became 13.8 Byr, the revision was within the margin of error.

    Type 1a have been the best "standard candles" to improve our estimates since they can be seen for distances far greater than anything else where we can have a reasonably good idea of their absolute magnitudes.

    Gravitational lensing may be improving other means in distance determination, but I'm not sure.

    I'm unclear what you are trying to demonstrate.
    Reply
  • rod
    Helio, the distance measurement for SN1987A does not show the universe true age is some 13.8 billion years old. The stellar parallax method of distance measurement in astronomy is the direct method using Earth's orbit as the baseline and stellar parallax does not show the *true age* of the universe, neither does angular size or expansion rates of various SN. Type 1a SN distances do not show the universe has been expanding for 13.8 billion years for example. In BB cosmology, the 13.8 billion years age of the universe is the Hubble time based upon H0. The Hubble time has a universe at least some 46 billion light years in radius, as seen from Earth. So far those distances cannot be verified, i.e. that the universe has such a large radius from Earth's frame of reference which is an important part of computing the Hubble time for the age of the universe using expansion. The Hubble time has a number of assumptions in the age calculation.

    There is a new report out showing the Hubble constant is 75.1 km/s/Mpc with age of the universe pegged at 12.6E+9 years old now. See, 'New approach refines the Hubble's constant and age of universe', https://phys.org/news/2020-07-approach-refines-hubble-constant-age.html. "Calculations drawn from observations of NASA's Wilkinson Microwave Anisotropy Probe in 2013 put the age of the universe at 13.77 billion years, which, for the moment, represents the standard model of Big Bang cosmology. The differing Hubble's constant values from the various techniques generally estimate the universe's age at between 12 billion and 14.5 billion years. The new study, based in part on observations made with the Spitzer Space Telescope, adds a new element to how calculations to reach Hubble's constant can be set, by introducing a purely empirical method, using direct observations, to determine the distance to galaxies, Schombert said."

    You asked, "I'm unclear what you are trying to demonstrate." What I demonstrated is that direct measurements for distances from Earth is limited and the Hubble time is not fixed based upon new values for H0 reported. as far as I am concerned, none of the *true age* claims tossed around are scientific fact. They may have some good arguments supporting, but areas that remain untested too, e.g. the radius of the universe from Earth out some 46 billion light years or more.
    Reply
  • Helio
    rod said:
    Helio, the distance measurement for SN1987A does not show the universe true age is some 13.8 billion years old.
    Right. The 1987a study is very similar to parallax measurements, which are angular. This takes distances to much greater distances than the limits you carefully noted.

    Type 1a SN distances do not show the universe has been expanding for 13.8 billion years for example.
    Agreed. But it helps tweak our methodology and accuracy, though how much it might help for a final age result is unclear.

    In BB cosmology, the 13.8 billion years age of the universe is the Hubble time based upon H0. The Hubble time has a universe at least some 46 billion light years in radius, as seen from Earth. So far those distances cannot be verified, i.e. that the universe has such a large radius from Earth's frame of reference which is an important part of computing the Hubble time for the age of the universe using expansion. The Hubble time has a number of assumptions in the age calculation.
    I think the CMBR helps verify it, but I could be wrong. I'm rusty at this, at best.

    The CMBR would have first taken place at a predicted temperature of about 3000K, IIRC. It was also required to have a near perfect Planck distribution (blackbody). This distribution was found but the redshift brought that initial blast down to a temperature of 2.73K.

    You asked, "I'm unclear what you are trying to demonstrate." What I demonstrated is that direct measurements for distances from Earth is limited and the Hubble time is not fixed based upon new values for H0 reported. as far as I am concerned, none of the *true age* claims tossed around are scientific fact. They may have some good arguments supporting, but areas that remain untested too, e.g. the radius of the universe from Earth out some 46 billion light years or more.
    Agreed. The final say could take decades. But what margin of error do you suggest the 13.8 Byr age should have?
    Reply
  • rod
    "But what margin of error do you suggest the 13.8 Byr age should have?" Helio, I offer no error bars here :) Just some observations about measuring the age of the universe via expansion rate (H0) and when the expansion is said to have begun.

    https://www.sciencedaily.com/releases/2020/07/200727114724.htm, New approach refines the Hubble's constant and age of universe "Summary: Using known distances of 50 galaxies from Earth to refine calculations in Hubble's constant, astronomers estimates the age of the universe at 12.6 billion years."

    According to globular clusters, the universe is 13.35 billion years old, https://phys.org/news/2020-07-globular-clusters-universe-billion-years.html
    Those GCs dated to 13.35 billion years old are needed in BB cosmology to map well with the CMBR age for the universe, e.g. Planck near 13.8 billion years ago. However, reports like what came out this week, show we could have GCs 13.35 billion years old, and a universe 12.6 billion years old in one some H0 measurements. This problem occurs in cosmology from time to time (finding objects in the universe older than the Hubble time for the BB event). The last I remember back in the 1990s, the ages of various GCs dated near or older than 14 billion years old conflicting with a 13.8 billion years old universe. Around and around cosmology dates for the age of the universe and objects found in the universe we seem to go (during Hubble and Einstein period the H0 value of 500 km/s/Mpc showed a universe < 2 billion years old) 😊---Rod
    Reply