Constraining Cosmology Using Large-scale Structures

Large-scale structures of the Universe provide a powerful probe to study the cosmic evolution history,including the structure formation process dominated by dark matter,and the expansion of the Universe due to dark energy.In this dissertation,we have revised the statistical methods for analysing large-scale structures from galaxy surveys.In particular,we focus on modelling and measuring a specific feature in the correlation functions of galaxies and cosmic voids,known as the Baryon Acoustic Oscillations(BAOs),which are used as a standard ruler to constrain the expansion history and geometry of the Universe.To have reliable estimates on the covariance of clustering measurements,large sets of synthetic catalogues are required,which mimic the observed distribution of tracers.We present two fast mock generation methods,PATCHY and EZMOCK,that both rely on perturbation theories combined with nonlinear and stochastic halo bias models,and reach high accuracies in terms two-point and three-point statistics.We have further studied the relation between halo mass and its environment,and found that halo mass depends not only on the local dark matter density,but also on non-local quantities such as the cosmic web environment and the halo-exclusion effect.Given these accurate relations,we have developed the HADRON-code(Halo mAss Distribution ReconstructiON)to assign halo masses to a spatial distribution of haloes(produced by PATCHY or EZMOCK).In particular,this method,followed by the Halo Abundance Matching technique,is applied for generating the thousands of PATCHY mock galaxy catalogues for the Baryon Oscillation Spectroscopic Survey(BOSS).In the second part of this dissertation,we have presented a novel definition and a parameter-free finder of cosmic voids,which are large under-dense regions spanning 10-100h-1Mpc.With a selection on void radii,we are able to identify troughs of the density field efficiently from galaxy surveys.With the help of PATCHY mocks,we unveil the BAO signature from cosmic voids for this first time both in simulations(about 10-?detection)and observations(>3-? detection using BOSS DR11 data).Then,we have explored a new parameterisation of the correlation function for a multi-tracer analysis with galaxies and cosmic voids,and developed a novel BAO model to measure the BAO peak scale.The joint sample in simulations show a>10%improvement on the BAO peak position constraint comparing to the result from galaxies alone after BAO reconstruction,which is equivalent with an enlargement of the survey volume by over 20%.Applying this method to the BOSS DR12 data,we have an 11%improvement for the low-z sample(0.2<z<0.5),but a worse constraint for the high-z sample(0.5<z<0.75),which is consistent with statistical fluctuations for the current survey volume.