Shot noise in multitracer constraints on fNL and relativistic projections: Power spectrum

Abstract

ABSTRACT Multiple tracers of the same surveyed volume can enhance the signal-to-noise on a measurement of local primordial non-Gaussianity and the relativistic projections. Increasing the number of tracers comparably increases the number of shot noise terms required to describe the stochasticity of the data. Although the shot noise is white on large scales, it is desirable to investigate the extent to which it can degrade constraints on the parameters of interest. In a multitracer analysis of the power spectrum, a marginalization over shot noise does not degrade the constraints on fNL by more than ∼30 per cent so long as haloes of mass $M\lesssim 10^{12}\, \mathrm{M}_\odot$ are resolved. However, ignoring cross shot noise terms induces large systematics on a measurement of fNL at redshift z < 1 when small mass haloes are resolved. These effects are less severe for the relativistic projections, especially for the dipole term. In the case of a low and high mass tracer, the optimal sample division maximizes the signal-to-noise on fNL and the projection effects simultaneously, reducing the errors to the level of ∼10 consecutive mass bins of equal number density. We also emphasize that the non-Poissonian noise corrections that arise from small-scale clustering effects cannot be measured with random dilutions of the data. Therefore, they must either be properly modelled or marginalized over.