A 120-Mpc periodicity in the three-dimensional distribution of galaxy superclusters

Abstract

According to the favour models for the formation of large-scale structure in the Universe (in which the dynamics of the Universe is dominated by cold dark matter), the distribution of galaxies and clusters of galaxies should be random on large scales. It therefore came as a surprise when a periodicity was reported 1 in the distribution of high-density regions of galaxies in the direction of Galactic poles, although the appearent lack of periodicity in other directions led to the initial report being regarded as a statistical anomaly 2 . A subsequent study 3−6 also claimed evidence for periodicity on the same scale, but the statistical significance of this result was uncertain due to small number of clusters used. Here, using a new compilation 7 of available data on galaxy clusters, we present evidence for a quasiregular three-dimensional network of rich superclusters and voids, with the regions of high density separated by ∼ 120 Mpc. If this reflects the distribution of all matter (luminous and dark), then there must exists some hithero unknown process that produces regular structure on large scales.During the past few years the number of clusters with measured redshifts has increased considerably. To search for the possible presence of a regularity of the distribution of matter in the Universe we have used a new compilation 7 of available data on rich clusters of galaxies catalogued by Abell and collaborators 8,9 . The compilation has made use of all (∼300) published references on redshifts of both individual galaxies and Abell galaxy clusters. Individual galaxies were associated with a given Abell cluster if they lay within a projected distance of ≤1.5 h −1 Mpc (1 Abell radius) and within a factor of two of the redshift estimated from the brightness of the cluster's 10-th brightest galaxy, using the photometric estimate of Peacock & West 10 (h is the Hubble constant in units of 100 km/s/Mpc). The compilation contains measured redshifts for 869 of the 1304 clusters with an estimated redshift up to z = 0.12. For the present analysis we used all rich clusters (richness class R ≥ 0) in this compilation with at least two galaxy redshifts measured. The omission of the 435 clusters without measured redshifts does not affect our result because an appropriate selection function was used.This cluster sample (including clusters with estimated redshifts) was used to construct a new catalogue of 220 superclusters of galaxies 11,12 . These are systems of clusters where the distances between nearest neighbours among member clusters do not exceed 1

Single field inflation and non-Gaussianity

“…To make the connection with the literature, note that we are not taking a 0 = 1 today. Rather, a 0 = 2ℓ H ≈ 6000 h −1 Mpc which implies that a preferred scale located at 0.004 h Mpc −1 [32] corresponds to a comoving k b = 24 in our case, while a preferred scale located at 0.052 h Mpc −1 [27] yields k b = 312 in our units. In order to illustrate the different forms of the initial spectrum, the functionh(k) is represented in Fig.…”

“…Using the CMBFAST code it was found [29] that this spike in the spectrum is seen at a ≥ 25 per cent smaller wavenumber than the second acoustic peak, while higher values of the baryon contribution Ω b may be needed to fit the amplitude of this feature. If we interpret this feature as originating from the primordial spectrum then, in order to be consistent with observations, the preferred scale k b must lie way below the horizon today, possibly at a scale corresponding to the turn of the power spectrum [27] or at the scale matching the first acoustic peak of the CMB temperature anisotropies [32]. One then sees that the presently available data already restricts the parameter space for the quantities k b and n [33].…”

Bumpy power spectra and ΔT/T

“…Broadhurst et al 1990;Einasto et al 1997;Landy et al 1996). For the concordance paradigm, use of a bump-like feature to account for this excess is not consistent with the constraints from the CMB data.…”

“…Broadhurst et al 1990;Einasto et al 1997;Landy et al 1996). Figure 5 shows our best-fitting bump model together with the standard ΛCDM model without a bump and the same model with a bump at k b = 0.05 h Mpc −1 , each model independently taking its optimal normalization to the full CMB data set.…”

“…Nevertheless, several independent data sets have provided such indications (See e.g. Broadhurst et al 1990;Einasto et al 1997;Landy et al 1996).…”