Evidence for primordial gravitational waves negated

On 17 March 2014, scientists working with the South Pole’s BICEP2 telescope announced that they had seen characteristic twisted patterns, called B modes, in the polarization of microwave photons coming from a significant patch of sky. The team, after accounting for contributions from dust in our galaxy, interpreted its observations as arising from primordial gravitational waves, stretched by cosmic inflation and imprinted on the cosmic microwave background (CMB; see Physics Today, May 2014, page 11 ). Several months later data from the Planck collaboration suggested that dust may have caused the BICEP2 result after all (see the Commentary by Mario Livio and Marc Kamionkowski, Physics Today, December 2014, page 8 ). Now a joint paper by researchers from BICEP2, the South Pole’s Keck Array collaboration, and Planck finds no solid evidence for primordial gravitational waves. At frequencies much above 200 GHz, the galactic-dust contribution to B modes dominates the gravitational-wave-induced CMB signal. The new joint work compared the B-mode distribution observed by Planck at 353 GHz, for which dust is surely the cause, with that observed by BICEP2, and later by Keck, at 150 GHz. The distributions were highly correlated, suggesting that dust is also responsible for the 150-GHz signal. The strength of gravitational-wave-induced CMB polarization is conventionally described by a dimensionless parameter, r. Last year’s BICEP2 announcement cited r = 0.2, several standard deviations away from zero; the new work bounds r to be less than 0.12. Evidence for gravitational waves may yet be lurking in the original BICEP2 data; if so, it will take more work to tease it out. (P. A. R. Ade et al., BICEP2/Keck and Planck collaborations, Phys. Rev. Lett. 114, 101301, 2015, doi:10.1103/PhysRevLett.114.101301 .)