A gravitational-lensing measurement of the Hubble constant

In 1929 Edwin Hubble confirmed that galaxies are receding from us with a speed proportional to their distance: v = H₀d. As late as the mid 1990s, the value of the proportionality constant H₀, the Hubble constant, was known only to be somewhere between 50 and 90 km/s per megaparsec (see the article by Mario Livio and Adam Riess, Physics Today, October 2013, page 41 ). With the help of space-based observatories, H₀ can now be determined with a precision of about 1%. The value obtained from a detailed map of the cosmic microwave background (CMB) is 66.93 ± 0.62 km/s/Mpc. But that determination is in tension with the value of 73.24 ± 1.74 km/s/Mpc derived from standard candles (Cepheid variables and type Ia supernovae, whose luminosities are known).

Now the H0LiCOW (H₀ Lenses in COSMOGRAIL’s Wellspring) collaboration has presented a comparably precise measurement based on its observations of three gravitationally lensed quasar systems. The H0LiCOW result, H₀ = 71.9 + 2.4 − 3.0 km/s/Mpc, agrees with the standard-candle determination, but it is about 2 standard deviations distant from the CMB-derived value.

When light traveling from a quasar to Earth passes by a sufficiently massive galaxy, the galaxy can act as a lens that bends the quasar light. As a result, Earthbound astronomers see multiple images of the quasar as shown in the figure. At times the brightness of the quasar flickers, and those fluctuations at the source are observed in the lensed images too. But since each image corresponds to a slightly different path length from quasar to telescope, the flickers appear at slightly different times for each image. The H0LiCOW team carefully measured those time delays, which are inversely proportional to H₀.

The Hubble constant determination from the CMB assumes, among other things, that the universe is flat and that dark energy is characterized by Einstein’s cosmological constant. If the conflicting values suggested by standard candles and lensed quasars hold up, some of the assumptions of cosmology’s now-standard model may need to be revised. (V. Bonvin et al., Mon. Not. R. Astron. Soc. 465, 4914, 2017, doi:10.1093/mnras/stw3006 .)