We use cookies to ensure that we give you the best experience on our website. By continuing to browse this repository, you give consent for essential cookies to be used. You can read more about our Privacy and Cookie Policy.

Durham e-Theses
You are in:

Search for non-Gaussianity in Large Scale Structure surveys

KARAGIANNIS, DIONYSIOS (2013) Search for non-Gaussianity in Large Scale Structure surveys. Masters thesis, Durham University.



In this work we put constraints on the primordial non-Gaussianities by using recent Large Scale Structure (LSS) surveys. The importance of measuring the amplitude of the primordial non-Gaussianity lies in the fact that it is the most prominent observational probe of the very early Universe. The plethora of the inflationary scenarios describing the early Universe makes it urgent to decide between them and create a solid physical theory for this era. The different inflation models predict different amount of non-Gaussianity in the primordial density perturbations, which will seed the LSS we observe. Therefore here we use the clustering results of prominent LSS surveys in order to test if they have the statistical power to constrain the primordial non-Gaussianity. We review the clustering of the radio sources from the NRAO VLA Sky Survey at z ∼ 1. The non-Gaussianity measured is one of the best determinations coming from LSS in the literature, f_{NL}= 62 ± 27 (68% CL). We also use the full scale range clustering of the LRG's from the CMASS of SDSS BOSS DR8 at z = 0.55. By using the scale dependence of the bias, originating from the existence of primordial non-Gaussianity, we fit non-Gaussian models to the large scales of our sample in order to measure the f_{NL}. The resulting fits show that there is room in this sample for non-Gaussianity. Although due to the large scale uncertainty errors the standard ΛCDM model cannot be excluded. Recently the measured non-Gaussianity from the SDSS BOSS CMASS sample, −92 < f_{NL} < 398 at 95% CL, shows that the constraints are not tight. This was expected because of the large scale statistical uncertainties in the clustering of this sample. Our best-fit measured f_{NL} = 71 ± 11 (1σ) is consistent with their measurements. The H-a emitters from HiZELS at a narrow redshift selection z = 2.23 are a promising survey for non-Gaussianity, but in order to gain any interesting constraints we have to wait for a larger sample. Finally we analyze the clustering of the ∼ 30, 000 quasar sample of SDSS BOSS DR9 at an effective redshift of z_{eff}= 2.4. The results show an amplitude excess in the clustering of the sample at the large scales. By fitting non-Gaussian models to the correlation function we measure, f_{NL} = 135 ± 9 at 1σ CL. ΛCDM fits the clustering results until 40 h-1 Mpc. However we cannot exclude the standard cosmological model since at the large scales that constrain f_{NL}, our results remain sensitive to the effects of systematic errors. We check the quasar sample for any potential systematics and particularly for the systematic effects of galactic extinction, seeing, sky brightness and foreground stars. Similar to previous studies the largest systematic comes from the presence of foreground stars. When we correct for such systematics we find, f_{NL}= 63 ± 16 (1σ). The measured amount of non-Gaussianity after correcting for the systematic effects is consistent with the results coming from the NVSS radio sources sample. Since the large scale amplitude of the clustering results is directly affected by systematics, we need to apply a more sophisticated method for correcting such effects. In any case, our original results show that the quasar sample shows excellent potential for determining the amplitude of primordial non-Gaussianity.

Item Type:Thesis (Masters)
Award:Master of Science
Keywords:Cosmology, non-Gaussianity, Large Scale Structures, clustering
Faculty and Department:Faculty of Science > Physics, Department of
Thesis Date:2013
Copyright:Copyright of this thesis is held by the author
Deposited On:26 Apr 2013 14:58

Social bookmarking: del.icio.usConnoteaBibSonomyCiteULikeFacebookTwitter