Covariance of cross-correlations: towards efficient measures for large-scale structure

We study the covariance of the cross-power spectrum of different tracers for the large-scale structure. We develop the counts-in-cells framework for the multi-tracer approach, and use this to derive expressions for the full non-Gaussian covariance matrix. We show, that for the usual auto-power statistic, besides the off-diagonal covariance generated through gravitational mode-coupling, the discreteness of the tracers and their associated sampling distribution can generate strong off-diagonal covariance, and that this becomes the dominant source of covariance as k>>k_f=2 pi/L. On comparison with the derived expressions for the cross-power covariance, we show that the off-diagonal terms can be suppressed, if one cross-correlates a high tracer-density sample with a low one. Taking the effective estimator efficiency to be proportional to the signal-to-noise ratio (SN), we show that, to probe clustering as a function of physical properties of the sample, i.e. cluster mass or galaxy luminosity, then the cross-power approach can out perform the auto-power one by factors of a few. We confront the theory with measurements of the mass-mass, halo-mass, and halo-halo power spectra from a large ensemble of N-body simulations. We show that there is a significant SN advantage to be gained from using the cross-power approach when studying the bias of rare haloes. The analysis is repeated in configuration space and again SN improvement is found. We estimate the covariance matrix for these samples, and find strong off-diagonal contributions. The covariance depends on halo mass, with higher mass samples having stronger covariance. In agreement with theory, we show that the covariance is suppressed for the cross-power. This work points the way towards improved estimators for clustering studies.