Optimal constraints on local primordial non-Gaussianity from the two-point statistics of large-scale structure

摘要: One of the main signatures primordial non-Gaussianity local type is a scale-dependent correction to bias large-scale structure tracers such as galaxies or clusters, whose amplitude depends on itself. The dominant source noise in power spectrum caused by sampling variance large scales (where non-Gaussian signal strongest) and shot arising from their discrete nature. Recent work has argued that one can avoid comparing multiple different bias, suppress optimally weighting halos mass. Here we combine these ideas investigate how well fluctuations potential be extracted two-point correlations dark matter. On basis $N$-body simulations with initial conditions halo catalogs perform Fisher matrix analysis statistics. Compared standard analysis, optimal multiple-tracer techniques applied yield up 1 order magnitude improvements ${f}_{\mathrm{NL}}$-constraints, even if underlying matter density field not known. In this case needs resolve all down ${10}^{10}{h}^{\ensuremath{-}1}{\mathrm{M}}_{\ensuremath{\bigodot}}$ at $z=0$, while already achieved mass threshold ${10}^{12}{h}^{\ensuremath{-}1}{\mathrm{M}}_{\ensuremath{\bigodot}}$. We compare our numerical results model find satisfactory agreement. Forecasting ${f}_{\mathrm{NL}}$-constraints methods when existing future survey data, $50{h}^{\ensuremath{-}3}\text{ }\text{ }{\mathrm{Gpc}}^{3}$ volume resolving ${10}^{11}{h}^{\ensuremath{-}1}{\mathrm{M}}_{\ensuremath{\bigodot}}$ $z=1$ will able obtain ${\ensuremath{\sigma}}_{{f}_{\mathrm{NL}}}\ensuremath{\sim}1$ (68% cl), factor $\ensuremath{\sim}20$ improvement over current limits. Decreasing minimum resolved halos, increasing obtaining maps further improve limits, potentially reaching level ${\ensuremath{\sigma}}_{{f}_{\mathrm{NL}}}\ensuremath{\sim}0.1$. This precision opens possibility distinguish types probe inflationary physics very early Universe.