Introduction The Copernican problem Methodology Results The Local Group factory and Bayesian inference Conclusions Cosmic flows, Bayesian Inference & the Local Universe Yehuda Hoffman (Racah Inst. of Physics, Hebrew University) XIIth Rencontres du Vietnam Large Scale Structure and Galaxy Flows July, 3rd - 9th, 2016, Quy Nhon, Vietnam 1 / 34 Introduction The Copernican problem Methodology Results The Local Group factory and Bayesian inference Conclusions Collaborators (partial list) Potsdam: Jenny Sorce, Stefan Gottloeber, Noam Libeskind, Matthias Steinmetz Madrid: Chris Brook, Alexander Knebe, Gustavo Yepes Lyon: Helene Courtois, Romain Graziani Saclay: Daniel Pomarede Hawaii: Brent Tully Copenhagen: Arianna di Cintio Moscow: Sergey Pilipenko Tartu Obser.: Elmo Tempel Jerusalem: Edoardo Carlesi 2 / 34 Introduction The Copernican problem Methodology Results The Local Group factory and Bayesian inference Conclusions Outline Cosmology with cosmic flows Cosmicflows Bayesian inference Near-field cosmology Constrained Local UniversE Simulations (CLUES)a ahttp://www.clues-project.org 3 / 34 Introduction The Copernican problem Methodology Results The Local Group factory and Bayesian inference Conclusions CLUES (in a nutshell) The Three Pillars of CLUES Reconstruction of the Large Scale Structure (LSS) from noisy, sparse and incomplete data Time machine: from the (present epoch) reconstructed LSS to initial conditions (ICs) Constrained simulations: from ICs to the present epoch nearby universe 4 / 34 Introduction The Copernican problem Methodology Results The Local Group factory and Bayesian inference Conclusions Near-field cosmology (Bland-Hawthorn, 1999) Near Field Cosmology Our local neighbourhood is the part of the universe that we know the best - faint galaxies, satellites, tidal streams, ... ΛCDM - is in an excellent agreement with the universe at large (scales), but tension exists on small scales: ’cusp vs. core’, abundance of satellites The value of H0: discrepancy between the near- and the far-field estimated values Nature of dark matter Galaxy formation The notion of ‘NEAR-field’ is stretched here to cover everything that is LOCAL. 5 / 34 Introduction The Copernican problem Methodology Results The Local Group factory and Bayesian inference Conclusions Phases vs. Power In the standard model of cosmology the universe emerges out of a Gaussian random perturbation field. 1 Z δ(r) = d3kδ exp ı k · r (2π)3/2 k δk = δk exp ı φk Gaussian random perturbation field → RANDOM PHASE APPROXIMATION Cosmologists are traditionally interested in the power (spectrum) of the perturbation field, P(k). Study of the near field involves the (reconstruction of the) phase structure. Here we focus on the recovery of the power & phases (of the near-field). 6 / 34 Introduction The Copernican problem Methodology Results The Local Group factory and Bayesian inference Conclusions The Copernican problem: near-field vs cosmology How [a]typical is the Local Group? How representative is the near field? The physicist says: I want it to be typical so I can practice [near-field] cosmology on it. The astronomer/astrophysicist says: I want it to be atypical. It is more interesting this way. The LG and the near field can be typical with respect to some properties and unique with respect to others. And it is a matter of degree - not black or white. 7 / 34 Introduction The Copernican problem Methodology Results The Local Group factory and Bayesian inference Conclusions Bayesian framework: WF/CRs In the Bayesian approach one is interested in the posterior probability of a model given observational data: P(model | data) ∝ P(data | model) P(model) P(model) is the prior probability (knowledge) of the model P(data | model) is the likelihood of the data given the (prior) model P(model | data) is the posterior probability Model: Gaussian random field, with the ΛCDM power spectrum Data: peculiar velocities of galaxies (Cosmicflows database) Reconstruction of the large scale structure by means of the Wiener filter (WF) and constrained realizations (CRs) 8 / 34 Introduction The Copernican problem Methodology Results The Local Group factory and Bayesian inference Conclusions Data: Cosmicflows-2 (CF2) The Astronomical Journal,146:86(25pp),2013October doi:10.1088/0004-6256/146/4/86 C 2013. The American Astronomical Society. All rights reserved. Printed in the U.S.A. ⃝ COSMICFLOWS-2: THE DATA R. Brent Tully1,Hel´ ene` M. Courtois1,2,AndrewE.Dolphin3,J.RichardFisher4,PhilippeHeraudeau´ 5, Bradley A. Jacobs1,IgorD.Karachentsev6,DmitryMakarov6,LidiaMakarova6, Sofia Mitronova6,LucaRizzi7,EdwardJ.Shaya8,JennyG.Sorce2, and Po-Feng Wu1 1 Institute for Astronomy, University of Hawaii, 2680 Woodlawn Drive, HI 96822, USA 2 Universite´ Claude Bernard Lyon I, Institut de Physique Nucleaire, F-69622 Lyon, France 3 Raytheon Company, 1151 E Hermans Rd., Tucson, AZ 85756, USA 4 National Radio Astronomy Observatory, 520 Edgemont Road, Charlottesville, VA 22903, USA 5 Argelander-Institut fur¨ Astronomie, Auf dem Hugel¨ 71, D-53121 Bonn, Germany Number of data The Astronomical Journal,146:86(25pp),2013October 6 Special Astrophysical Observatory, N Arkhyz, KChR 369167, Russia Tully et al. 7 W. M. Keck Observatory, 65-1120 Mamalahoa Highway, Waimea, HI 96743, USA 8 Department of Astronomy, University of Maryland, College Park, MD 20742, USA Received 2013 May 31; accepted 2013 July 25; published 2013 September 5 points: ≈8000 1200 ABSTRACT Cosmicflows-2 is a compilation of distances and peculiar velocities for over 8000 galaxies. Numerically the largest contributions come from the luminosity–line width correlation for spirals, the Tully–Fisher relation (TFR), and (4814 grouped) 1000the related fundamentalCF1 plane relation for E/S0 systems,CF2 but over 1000 distances are contributed by methods that provide more accurate individual distances: Cepheid, tip of the red giant branch (TRGB), surface brightness fluctuation, Type Ia supernova, and several miscellaneous but accurate procedures. Our collaboration is making important contributions to two of these inputs: TRGB and TFR. A large body of new distance material is presented. In addition, an effort is made to ensure that all the contributions,SFI++ both our own and those from the literature, are on the 800 1 1 same scale. Overall, the distances are found to be compatible with a Hubble constant H0 74.4 3.0kms− Mpc− . The great interest going forward with this data set will be with velocity field studies. Cosmicflows-2= ± is characterized Median redshift: by a great density and high accuracy of distance measures locally, falling to sparse and coarse sampling extending to z 0.1. N 600= Key words: catalogs – galaxies: distances and redshifts −1 Online-only material: color figures, machine-readable and VO tables 5895 km s 400 1. INTRODUCTION (Cepheid PLR). There have been recent modest refinements to this scale (Riess et al. 2011;Freedmanetal.2012). This paper presents a compendium of galaxy distances Cepheid variables are young stars, frequently in regions of and peculiar200 velocities that is being called Cosmicflows-2.A obscuration, and only present in galaxies currently forming stars. precursor catalog by Tully et al. (2008)hasretroactivelybeen Our second route to a zero point calibration strictly involves old called Cosmicflows-1.Inbothcases,thecomponentsofthe stars. Distances are determined from the luminosities of red catalogs are a0 mix of new material and information from the giant branch (RGB) stars at the onset of core helium burning, at alocationinastellarcolor–magnitudediagram(CMD)known literature. Cosmicflows-1 provided, at the time, the3 densest 4 4 coverage of distances locally, but it was severely restricted by as the tip of the red giant branch (TRGB). For this discussion, 0 1 5.0•10 we avail1.0•10 of photometry obtained1.5•10 with HST with filters that acutoffof3000kms− .ThenewCosmicflows-2 enhances the -1 density of coverage locally and extends coverage sparselyV to (kmapproximate s ) I band. There is now considerable experience 1 CMB 30,000 km s− . that demonstrates the viability of TRGB measurements in this The current compilation draws on distance determinations band (Lee et al. 1993;Makarovetal.2006;Madoreetal. by six distinct1200 methods. In four cases, all the base material 2009;Dalcantonetal.2009). The zero point is provided is drawn from the literature and the endeavor here has been by bootstrapping from observations of spheroidal companions to ensure a uniform scaling. In two important cases, our to the Milky Way that link TRGB and horizontal branch collaboration has made major observational contributions, with magnitudes. The horizontal branch magnitudes are fixed to an much of1000 our material being releasedCF2 for the first time with this absolute scale through studies of globular clusters. Rizzi et al. publication. (2007)demonstratedthatthisPopulationIIpathtoacalibration The six independent methodologies have distinct merits and gives extragalactic distances that agree with the Cepheid PLR deficiencies. The samples are now getting sufficiently large scale at the level of 0.01 mag. Further confirmation has come because of800 the six procedures,SFI++ so each has good overlap with the from the agreement in measured distances to NGC 4258 from 9 / 34 others. Multiple measurements of individual galaxies and within maser observations, the Cepheid PLR, and TRGB (Herrnstein groups and clusters are creating an increasingly tight lattice of et al. 1999; Rizzi et al. 2007; Mager et al. 2008; Riess et al. 2011). distances on a common scale. Essentially all galaxies