Gravitational waves could resolve Hubble constant debate

22 Feb 2019 Alex Petkov

Gravitational waves from a neutron star collision (Courtesy: NASA's Goddard Space Flight Center/CI Lab)

Simulations by an international team of researchers have shown that new measurements of gravitational waves could finally resolve the discrepancy in Hubble’s constant reported using different measurement techniques. Accumulating gravitational-wave signals from the mergers of 50 binary neutron stars, the scientists found, will yield the most accurate value of the constant to date – which would not only settle the debate but also confirm whether there are issues with the current standard cosmological model.

The Hubble constant represents the rate at which the universe is currently expanding and is vital for calculating both its age and its size. The constant is also widely used in astronomy to help determine the masses and luminosities of stars, the size scales of galaxy clusters, and much more besides. However, two different techniques for estimating the value of Hubble’s constant have yielded very different results

To measure Hubble’s constant directly, scientists need to know a galaxy’s outward radial velocity and its distance from the Earth. The first of these measurements can be obtained from the galaxy’s spectroscopic redshift, but the distance to the galaxy is more difficult to determine directly.

A common way of estimating distance is to exploit so-called “standard candles” – Cepheid variable stars or type 1a supernovae that have known absolute luminosities. In 2016 the best estimate for the Hubble constant obtained this way was 73.2 km s–1 Mpc–1 – vastly different from the value of 67.8 km s–1 Mpc–1 obtained in the same year by studying the radiation of the Cosmic Microwave Background (CMB). The discrepancy is yet to be explained since the values should agree if the standard cosmological model is correct.

In this case, the scientists implemented PPD to simulate measurements of the Hubble constant using these two different methods, and to check their consistency with the standard cosmological model. One interesting finding is that there’s at least a 6% chance that the current discrepancy in the Hubble constant is purely due to random error. In this new study, researchers from Europe and the US attempted to reconcile these two results. The scientists exploited the concept of “posterior predictive distribution” (PPD), a methodology often used to determine the reproducibility of experimental results. PPD relies on a dynamic view of probability – in other words, one that changes as new information is obtained.

They then simulated how new independent data could help resolve the debate. Gravitational waves from merging neutron stars seemed a promising avenue to explore, since their signal yields constraints on the distance to the binary stars. Measurements of gravitational waves should, therefore, provide an estimate of the Hubble constant without making any assumptions about the cosmology of the Universe.

the researchers found that 50 detections of gravitational-wave signals from merging neutron stars would be needed to properly arbitrate between the two different values for the Hubble constant. Including such a dataset within their PPD simulations would, they claim, yield the most accurate value of the Hubble’s constant yet measured – with an error of below 1.8%. Judging by current progress, observations of those 50 neutron-star mergers could well be achieved within the next decade.

Full results are published in Physical Review Letters.

physicsworld.com 22/2/2019