The galaxy power spectrum: precision cosmology from large-scale structure?

Published galaxy power spectra from the two-degree field galaxy redshift survey (2dFGRS) and Sloan Digital Sky Survey (SDSS) are not in good agreement. We revisit this issue by analysing both the 2dFGRS and SDSS Data Release 5 (DR5) catalogues using essentially identical techniques. We confirm that the 2dFGRS exhibits relatively more large-scale power than the SDSS, or, equivalently, SDSS has more small-scale power. We demonstrate that this difference is due to the r-band selected SDSS catalogue being dominated by more strongly clustered red galaxies, which have a stronger scale-dependent bias. The power spectra of galaxies of the same rest-frame colours from the two surveys match well. If not accounted for, the difference between the SDSS and 2dFGRS power spectra causes a bias in the obtained constraints on cosmological parameters which is larger than the uncertainty with which they are determined. We also found that the correction developed by Cole et al. to model the distortion in the shape of the power spectrum due to non-linear evolution and scale-dependent bias is not able to reconcile the constraints obtained from the 2dFGRS and SDSS power spectra. Intriguingly, the model is able to describe the differences between the 2dFGRS and the much more strongly clustered Luminous Red Galaxy (LRG) sample, which exhibits greater non-linearities. This shows that more work is needed to understand the relation between the galaxy power spectrum and the linear perturbation theory prediction for the power spectrum of matter fluctuations. It is therefore important to accurately model these effects to get precise estimates of cosmological parameters from these power spectra and from future galaxy surveys like Pan-STARRS, or the Dark Energy Survey, which will use selection criteria similar to the one of SDSS.