I was more convinced that the original paper may have problems by https://arxiv.org/abs/2010.09761 From that preprint:

> We find that the 12th-order polynomial fit to the spectral passband utilised in the published study leads to spurious results.

Using a 12th-order polynomial fit is very suspicious? And I think that it is very interesting to see Figure 2, top-left. The signal is very noisy and the polynomial fit is not convincing. That graph is a very big red flag.

Something is terribly wrong if astronomy of all fields encountered a reproducibility crisis.