Non-linear corrections to the cosmological matter power spectrum and scale-dependentgalaxy bias: implications for parameter estimation

Abstract

We explore and compare the performances of two non-linear correctionand scale-dependent biasing models for the extraction of cosmologicalinformation from galaxy power spectrum data, especially in the context ofbeyond-ΛCDM (CDM: cold dark matter) cosmologies. The first model is the well knownQ model, first applied in the analysis of Two-degree Field Galaxy Redshift Survey data. The second,the P model, is inspired by the halo model, in which non-linear evolution and scale-dependent biasing areencapsulated in a single non-Poisson shot noise term. We find that while the two models performequally well in providing adequate correction for a range of galaxy clustering data in standardΛCDM cosmology and in extensions with massive neutrinos, theQ model can give unphysical results in cosmologies containing a subdominantfree-streaming dark matter whose temperature depends on the particle mass, e.g., relicthermal axions, unless a suitable prior is imposed on the correction parameter.This last case also exposes the danger of analytic marginalization, a techniquesometimes used in the marginalization of nuisance parameters. In contrast, theP model suffers no undesirable effects, and is the recommended non-linear correction modelalso because of its physical transparency.