Removing the giants and learning from the crowd: a new SZ power spectrum method and revised Compton $y$-map analysis

The Sunyaev-Zeldovich (SZ) effect provides a powerful cosmological probe, which traditionally is approached independently as cluster number count (CNC) or power spectrum (PS) analysis. Here, we devise a new method for analysing the $y$-map by introducing the survey completeness function, conventionally only used in the CNC analysis, in the $yy$-PS modeling. This provides a systematic method, based mainly on SZ observables, for obtaining two complementary $y$-maps, one incorporating detected/resolved clusters and the other relying only on diffuse/unresolved SZ contributions. We use the catalogue of clusters obtained in the \Planck CNC analysis to define the completeness function linking these two $y$-maps. The split depends on the chosen signal-to-noise detection threshold, which we vary in our discussion. We carefully propagate the effect of completeness cuts on the non-Gaussian error contributions in the $yy$-PS analysis, highlighting the benefits of masking massive clusters. Our analysis of the \Planck $yy$-PS for the unresolved component yields a mass bias of $b=0.15\pm0.04$, consistent with the standard value ($b\approx0.2$), in comparison to $b=0.4\pm 0.05$ for the total $yy$-PS. We find indications for this drift being driven by the CIB-tSZ cross correlation, which dominantly originates from clusters in the resolved component of the $y$-map. Another possible explanation is the presence of a mass-dependent bias, which has been theoretically motivated and can be quantified with our novel method. We furthermore find first hints for the presence of the 2-halo terms in the $yy$-PS. Finally, the proposed method provides a new framework for combining the complementary information of the CNC and PS analyses in upcoming SZ surveys.