“We hope that interested people will repeat our calculations and will make up their own minds regarding the significance of the results. It is obvious that “belief” is never an alternative to “understanding” in physics.” -J. Creswell et al.
Three times now, the LIGO collaboration has produced very strong evidence that black hole pairs, from across the Universe, inspiraled and merged, producing gravitational waves. The twin LIGO detectors in Hanford, WA and Livingston, LA each detected these signals, and the signals were correlated between both detectors. For the first time ever (and the second, and the third), we had directly detected gravitational waves.
But last month, a team of independent scientists from Denmark attempted to reproduce LIGO’s analysis, and noticed something that shouldn’t be there: noise correlations between the two detectors. Noise is supposed to be uncorrelated, and yet the noise correlations peaked at the moment of the inspiral-and-merger event. It doesn’t mean that gravitational waves aren’t real, but it does mean that LIGO, perhaps, has room for improvement.