!Licia!Verde! ICREA & ICC-UB Barcelona University of Oslo, Norway Cosmology and fundamental Physics! Some!examples!
http://icc.ub.edu/~liciaverde Context • Cosmology over the past 20 years has made the transition to precision cosmology • Cosmology has moved from a data-starved science to a data-driven science • Cosmology has now a standard model. The standard cosmological model only needs few parameters to describe origin composition and evolution of the Universe • Deep links to fundamental physics (e.g.,SM, BSM) • Big difference between modeling and understanding
AIM:!!propose!some!open!ques8ons!and!s8umulate!discussion! The!era!of!precision!cosmology:! LCDM: the standard model for cosmology Few parameters describe the Universe composition and evolution
Homogenous background Perturbations What s!next?!! Look!for!devia8ons!from!the!standard!model!
Test physics on which it is based and beyond it
• Dark!energy! • Nature!of!ini8al!condi8ons:!Adiaba8city,! Gaussianity! • Neutrino!proper8es!! • Infla8on!proper8es! • Beyond!the!standard!model!physics…! Cosmology!is!special!
We!can’t!make!experiments,!only!!observa8ons! The curse of cosmology We only have one observable universe
We can only make observations (and only of the observable Universe) not experiments: we fit models (i.e. constrain numerical values of parameters) to the observations: Any statement is model dependent
Gastrophysics and non-linearities get in the way : Different observations are more or less “trustable”, it is however somewhat a question of personal taste (think about Standard & Poor’s credit rating for countries)
Results will depend on the data you (are willing to) consider. (robustness?) Not all observations are created equal
….And the Blessing We can observe all there is to see We only have one observable universe ….And the Blessing
We can observe all there is to see
And almost do
We only have one observable universe challenges! From!precision!cosmology!to!accurate!cosmology! (Peebles!2002)!
As!the!sta8s8cal!errors!shrink…..! ! ! Systema8c!errors!must!be!kept!under!exquisite!control!!
!They!!sneak!in!from!everywhere:!! !Theory,!Astrophysics,!Observa8ons,!observables,!etc…!
There!is!no!systema8c!way!to!address!systema8c!errors! Dark!maWer!
• Evidences!for!dark!maWer!come!(Only?)!from! the!sky! • Indica8on!of!physics!BSM! Rota8on!curves!of!galaxies! CMB!and!perturba8ons! BAO!
P/Psmooth!
Anderson et al. 2014 Not!forge[ng!the!bullet!cluster! You!could!try!to!avoid!it….! ! But!you!would!have!to!give!up!GR! Neutrino!proper8es!
Neutrinos contribute at least to ~0.5% of the total matter density
Use!the!en8re!Universe!as!“detector”!!
What!is!a!neutrino!for!cosmology?!
• Behaves like radiation at T~ eV (recombination/decoupling) • Eventually (possibly) becomes non-relativistic, behaves like matter • Small interactions (not perfect fluid) • Has a high velocity dispersion (is “HOT”) Cosmic Neutrino Background
A relict of the big bang, similar to the CMB except that the CvB decouples from matter after 2s (~ MeV) not 380,000 years
At decoupling they are still relativistic (mν << Τν) ! # large velocity dispersions (1eV ~ 100 Km/s) Recall: T~1eV Matter-radiation equality, T=0.26eV at recombination
600Billion nu/s/cm3 from the sun ~100nu/cm3 from CvB The cosmic neutrino background has been detected at >> 4 σ#
WMAP SPT 99.7% or 0.003 Relict neutrinos influence in cosmology
Primordial nucleosynthesis CMB Large-scale structure
& H0
T Total mass >~1 eV become non relativistic before recombination CMB Total mass <~1 eV become non relativistic after recombination: alters matter-radn equality but effect can be cancelled CMB by other parameters Degeneracy After recombination =0) FINITE NEUTRINO MASSES ν Σm = 0 eV SUPPRESS THE MATTER POWER k,m ( SPECTRUM ON SCALES SMALLER P Σm = 0.3 eV )/ THAN THE FREE-STREAMING k ( LENGTH P m = 1 eV linear theory Σ Different masses become non-relativistic a slightly different times Cosmology can yield information about neutrino mass hierarchy Cosmology is the key to determine the absolute mass upper upper } KATRIN Σ<0.3 (95%CL) in a minimal LCDM scenario Until recently (pre-Planck) was the consensus Beware of systematics!!!!! It would be of great value to have an internal consistency check (more later) ) Σm = 0 eV =0 ν Σm = 0.3 eV k,m ( P )/ Σm = 1 eV k ( linear theory P Neutrinos!effect!on!the!maWer!P(k)! Σ (total mass) Note that non-linearities enhance the signal This is for MATTER Current constraints:Σ<0.3eV in real space The future is bright! Wagner, LV, Jimenez, 2012 What about real world effects? • Baryonic physics (lensing and galaxy surveys) • Bias (galaxy surveys) • redshift space (galaxy surveys) CMB back to the rescue Lensing From Hu Okamoto 2001 T E T E B CMB back to the rescue Lensing Planck paper XVII Infancy… but promising Neff: number of effective species 2 8⇡G 4 H (t) (⇢ + ⇢ ) ⇢ T N Standard: Neff=3.045 ' 3 ⌫ ⌫ / e↵ Extra radiation, boosted expansion rate Any thermal background of light particles, anything affecting expansion rate Systematics! Look at BBN Neff around 3 to 4 Look at CMB: effects matter-radn equality and so sound horizon at decoupling -> degeneracy with ωm and H Anisotropic stress, zeq on diffusion damping Main effect: increasing Neff increases Silk Damping scale (for fixed θs), small phase shifts too Hou et al 2011 Cosmological analyses consistently give best fit values >3.04. “dark radiation” But analyses are NOT independent (WMAP is always in common, H0 many times in common) See also the white paper Abazajian et al. arXiv:1204.5379, Claims of non-zero neutrino mass detection since march 2013 arxiv:1307.7715 arxiv:1308.5870 arxiv:1308.3255 Cosmic!Concordance?! Nonazero!sterile!neutrino!mass!only!favoured!due!to:! ! tension!between!CMB!and!clusters!(Planck!SZ,!Xaray)!in!σ8–Ωm!plane!! ! degeneracy!between!σ8!&!neutrino!mass.! ! Leistedt,!Peiris,!Verde,!2014! Hierarchy effect on the shape of the linear matter power spectrum Δ# Neutrinos of different masses have different transition NH IH redshifts from relativistic to non-relativistic behavior, and their individual masses and their mass splitting change the details m Δ sol of the radiation-domination to matter- domination regime.! Δmatmo Δmatmo Δmsol Jimenez, Kitching, Penya-Garay, Verde, JACP2010 Cosmology is (mostly) sensitive to |Δ| Still offers a powerful consistency check What would it take to measure Δ? Basically: the ultimate experiment In combination with v0ββ experiements can help answering: Are neutrinos Dirac or Majorana? Dirac!or!Majorana?!!!!!!!!!!hierarchy! Jimenez, Kitching, Penya-Garay, Verde, 2010 Complementarity:!cosmology,!par8cle!physics! Jimenez, Kitching, Penya-Garay, Verde, 2010, Wagner, LV, JimenezJimenez, 2012 Garay, Kitching, LV, 2010 Windows into the primordial Universe Recombination 380000 yrs Atomic physics/GR Nucleosynthesis 3 minutes Nuclear phyiscs LHC TeV energies inflation 10 -30 s (?) GUT? Big BANG Infla8on! Before!the!discovery!of!the!Higgs! !scalar!fields!in!physics!were!just!a!product!of!the!imagina8on…!! What mechanism generated the primordial perturbations? Inflation: Horizon problem Flatness problem Accelerated expansion: Structure Problem Quantum fluctuations get stretched to become classical and super-horizon The shape of the primordial power spectrum encloses information on the shape of the inflaton potential (CMB+Large scale structure) The energy scale of inflation is given by primordial tensor modes amplitude (CMB polarization!) Hints!to!UV!comple8on!of!gravity?! Inflationary predictions: or Inflation Checklist Simplest Inflationary Models • Spatially flat universe • Nearly Gaussian initial perturbations • Adiabatic initial conditions • Power spectrum spectral index nearly scale invariant (small red tilt in many implementations) • Super-horizon perturbations • A stochastic background of gravity waves Spectacular success Flat Small, homogeneous Adiabatic Gaussian ~Scale invariant Super-horizon Consistency with Simplest Inflationary Models Ωtot = 1.005 ± 0.0064 • Spatially flat universe (with lensing and BAO) • Nearly Gaussian initial perturbations fNL = 2.7 5.8 This means at better± than 0.01% • Adiabatic initial conditions • Power spectrum spectral index nearly scale invariant ns = 0.959 ± 0.007 (small red tilt in many implementations) • Super-horizon perturbations • A stochastic background of gravity waves Renewed confidence with Planck collaboration 2013 results Genera8on!of!CMB!polariza8on! • Temperature!quadrupole!at!the!surface!of! last!scaWer!generates!polariza8on.! From Wayne Hu! Potential hill Potential well Polariza8on!for!density!perturba8on!! • Radial!(tangen8al)!paWern!around!hot! (cold)!spots.! And it has been seen! Komatsu, WMAP7yrs team (2010) Theory prediction Observed Planck Measurement Cold spot Hot spot I Q I Q Cold spot Hot spot Theory prediction Planck collaboration 2013 Large-scale TE anti correlation Velocities (hot to cold) Density mode Hot due to doppler During decoupling Outside the horizon quadrupole is given by velocities Velocities are off-phase with density Large-scale TE anti correlation Causal Seed model (Durrer et al. 2002) Primordial Isocurvature i.c. WMAP TE data Peiris et al 2003 Primordial Adiabatic i.c. Super horizon Consistency with Simplest Inflationary Models • Spatially flat universe • Nearly Gaussian initial perturbations • Adiabatic initial conditions • Power spectrum spectral index nearly scale invariant (small red tilt in many implementations) • Super-horizon perturbations • A stochastic background of gravity waves Planck!polariza8on!(teasers)!plots! Planck!collabora8on!2013! Gravity Waves… are different Image from J. Rhul. Gravity Waves Image from J. Rhul. Polarization pattern! E and B modes polarization E polarization from scalar and tensor modes B polarization only from tensor modes “smoking gun” Kamionkowski, Kosowsky, Stebbings 1997, Zaldarriga & Seljak 1997 Why polarization? Temperature Shape of the inflation potential (and E-modes) Energy Scale of Inflation (Height of the potential) B-modes are needed! Tensor to scalar ratio, r * 0.1 *I will mention bicep later “Foregrounds are horrible but people are brave” (BICEP P.I.) The challenge T P Typical degree-scale brightness fluctuations (150GHz) Ground, Telescope mount etc 3-300 K 108- 1010 Atmosphere 30 mK - 3 K 106- 108 Galaxy 0.3-30mK 104- 106 CMB T anisotropies 30µK 103 Lensing B modes (at arcmin) 300 nK 10 r=0.01 B-modes 30 nK noise you want to reach <10 nK Adapted from C. Pryke Planck!limits!on!r:!How!they!work!! For!such!r!10%!of!the!TT!signal!at!low!l!comes!from!tensors!(too!much)! For!a!LCDM!power!law!power!spectrum!!the!!probability!p(r>=bicep)=0.0033,! !according!to!Planck!data! ! Monte Carlo simulation of the inflationary flow equations. BICEP!MEASUREMENT! Extraaordinary!claim…..! Fig!adapted!from!Verde!Peiris!Jimenez!2006! Why!extra!ordinary!claim?! It!implies:! There!is!a!stochas8c!background!of!gravity!waves!as!infla8on!predicts! !We!know!the!energy!scale!of!infla8on! We!know!how!much!the!field!has!moved!if!single!field! We!have!some!hope!to!test!the!consistency!rela8on!!nt!=a!8!r! ! ! Extra!ordinary!claims….! Need!extraaordinary!proofs! does!the!frequencyadependence!of!the!signal!look!like!CMB?! This!is!cri8cal!! ! Here:!difficult!game!to!play!! with!only!2!frequencies! (150!and!100!GHz)!one!of!them! with!high!noise!! This!plot!!says!that!it!could!be!synchrotron!! at!2!sigma…! !more!likely,!!dust.! Really,!we!need!more!frequencies! Planck! map!in!orthographic!projec8on!of!the!150!GHz!CBB"amplitudes!at!l!=!80,!computed!from!the! !Planck"353!GHz!data!,!es8mated!contamina8on!from!dust! Planck:!The!killer!plot! Consistency with Simplest Inflationary Models • Spatially flat universe (generic) • Nearly Gaussian initial perturbations • Adiabatic initial conditions • Power spectrum spectral index nearly scale invariant (small red tilt in many implementations) • Super-horizon perturbations (generic) • A stochastic background of gravity waves (this is actually generic) Gravity!and!GR!effects! !On!horizonascales!!the!Poisson!equa8on!gets!quadra8c!correc8ons:! Needs!IC!set!up!of!infla8on,!parallels!the!TE!an8acorrela8on.! P P / P P Verde & Matarrese 2009 At a potentially detectable level! Villa et al 2014 To!conclude! …!the!maximally!boring!universe…! The!standard!cosmological!model!has!survived!ever!more!stringent!tests! Devia8ons!from!it!!are!even!more!constrained! Precision cosmology -> address fundamental physics questions! (examples: neutrinos, initial conditions…), challenging!! Eventually!something!will!have!to!give,!the!model!IS!incomplete! The!point!is!how!much!smaller!would!the!observa8onal!error!bars!have!to!be! “We!can’t!live!in!a!state!of!perpetual!doubt,!so!we!make! up!the!best!story!possible!and!we!live!as!if!the!story!were! true.”! Daniel!Kahneman!about!theories! Challenges!! (or!what!keeps!me!up!at!night)! Cosmology!tends!to!rely!!hevily!!on!models!(both!for!“signal”!and!the!“noise”)! Essen8ally,!all!models!are!wrong!,!but!some!are!useful! (Box!and!Draper!1987)! What!can!be!measured!in!a!modelaindependent!way?! In!the!era!of!precision!and!accurate! cosmology…! There!is!no!room!for!spherical!cows!! END! Sta8s8cal!cosmology! From!observa8ons!of!the!observed(observable)!Universe!! we!want!to!make!inferences!about!the!proper8es!of…! the!en)re!Universe,!! seen!as!!an!ensamble!of!all!possible!observaed/able!Universes! of!which!our!observed!one!is!a!random!draw! “Fair!sample”! Cosmic!variance! No"wonder"that"many"cposmologists"are"Bayesian!" Any!theore8cal!model!will!mostly!!predict!! sta)s)cal!proper)es!of!the!Universe!! BAO! Observe photons Photons coupled to baryons See dark matter If baryons are ~1/6 of the dark matter these baryonic oscillations should leave some imprint in the dark matter distribution (gravity is the coupling) What mechanism generated the primordial perturbations? Inflation generates next to Gaussian perturbations: quantum fluctuations stretched to become classical by the expansion. The mechanism is similar to Hawkings radiation.Horizon (The event problem horizon being the complementary concept to the particle horizon…) Flatness problem The shape of the primordial power spectrum encloses information on the shape of the inflaton potential (CMB+Large scale structure) The spectrum of these perturbations will be ~ a power law: different models of inflation predict slightly different power laws, but all close to scale invariant .. The energy scale of inflation is given by primordial tensor modes amplitude (CMB polarization!) McMahon adapted by Peiris The Future is Bright Forecasts for Euclid Carbone et al 2011 Σ Detailed errors depend on what assumptions about underlying cosmology one is willing to make