Browsing by Author "Bengaly, Carlos"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Determining the observer’s velocity using radio continuum surveys
    (University of the Western Cape, 2019) Randriamiarinarivo, Nandrianina; Maartens, Roy; Bengaly, Carlos
    In the standard (‘concordance’) model of Cosmology, there is a fundamental assumption that the Universe is statistically isotropic and homogeneous on large scales, known as the Cosmological Principle. The Cosmological Principle requires that the dipole anisotropy apparent in the CMB should also be observed in galaxy number counts if this signal occurs due to the aberration and Doppler effects from our peculiar motion. This thesis will investigate the accuracy with which the cosmic kinematic dipole can be determined by comparing real data from NRAO VLA Sky Survey (NVSS) catalog with the simulated sky maps following its specifications. The mock maps are generated using FLASK code which assumes a lognormal distribution for the radio count density field from z=0 to z =4 and taking as an entry an angular power spectrum from CAMB which assumed a flat ΛCDM cosmology and a redshift distribution. After analising the kinematic dipole, we turn to the analysis of statistical isotropy in the catalog. We used ANalysis Of Variance (ANOVA) test on patches in the sky of different radii as one of the statistical tools for the analysis. We found that as we go to a higher radius for the patches, we have a better agreement between the theory and the observation as expected. We also saw that the more we are rigorous on the rejection criteria, the smaller is the discrepancy between the observed and simulated number count distribution in the sky. We found an optimum choice of 25◦ as patch size, and if the accepted patches have a maximum of 30% of their pixels masked. Therefore, we find that the NVSS data agrees with the fundamental assumption of statistical isotropy at angular scales > 20◦.
  • Thumbnail Image
    Item
    Null tests of the concordance model in the era of Euclid and the SKA
    (Elsevier, 2021) Bengaly, Carlos; Clarkson, Chris; Kunz, Martin; Maartens, Roy
    We perform null tests of the concordance model, using H (z) measurements that mimic next-generation surveys such as Euclid and the SKA. To this end, we deploy a non-parametric method, so that we make minimal assumptions about the fiducial cosmology as well as the statistical analysis. We produce simulations assuming different cosmological models in order to verify how well we can distinguish between their signatures. We find that SKA- and Euclid-like surveys should be able to discriminate sharply between the concordance and alternative dark energy models that are compatible with the Planck CMB data. We conclude that SKA and Euclid will be able to falsify the concordance model in a statistically significant way, if one of the benchmarks models represents the true Universe, without making assumptions about the underlying cosmology.
  • Thumbnail Image
    Item
    Testing the statistical isotropy of the universe using radio survey data
    (University of the Western Cape, 2019) Baloyi, Mathobela Albert; Maartens, Roy; Bengaly, Carlos
    The Cosmological Principle forms part of one of the most fundamental hypotheses of modern Cosmology. So it is very important to assess whether it holds true using observational data, or whether it consists of a mathematical simplification. We probe the statistical isotropy of the Universe using the existing radio continuum data, by means of a local variance estimator. In order to investigate this, we analyse the number count variance of the radio catalog by looking at patches of approximately 10, 15, 20 & 25 degrees in radii, and thus comparing it to mock catalogs which reproduce the matter density power spectrum, as well as the same sky coverage of the real data. We establish criteria for accepting patches that have more than 90%, 70% & 50% of their pixels not masked. We make use of the NRAO VLA Sky Survey (NVSS), whose operational frequency is 1.4 GHz. We perform statistical tests for detecting possible departures from statistical isotropy using galaxy number counts with flux limits of 20 < SNVSS < 1000 mJy. We also compare the real data to the mock catalogs of the radio data in order to assess the statistical significance of our results. We use the local variance estimator for testing the statistical isotropy of our data sample. We find that the statistical properties of our sample are in reasonable agreement with the standard cosmological model. The mean of the distribution for the data falls well within the 95% confidence interval of the average of the simulated mocks. For all the radii and acceptance criteria for the patches, we found no significant deviations beyond those allowed by the standard model. As expected there were no large discrepancies between our mocks and the data. The results are consistent with statistical isotropy.