49
Wim de Boer, Karlsruhe Kosmologie VL, 13.12.2012 1 Einteilung der VL 1. Einführung 2. Hubblesche Gesetz 3. Antigravitation 4. Gravitation 5. Entwicklung des Universums 6. Temperaturentwicklung 7. Kosmische Hintergrundstrahlung 8. CMB kombiniert mit SN1a 9. Strukturbildung 10. Neutrinos 11. Grand Unified Theories 12.-13 Suche nach DM HEUTE

Einteilung der VL - KITdeboer/html/Lehre/Kosmo... · 2012. 12. 13. · Lineweaver 1997 peak trough Sky Maps Power Spectra We “see” the CMB sound as waves on the sky. Use special

  • Upload
    others

  • View
    1

  • Download
    0

Embed Size (px)

Citation preview

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 1

    Einteilung der VL

    1. Einführung

    2. Hubblesche Gesetz

    3. Antigravitation

    4. Gravitation

    5. Entwicklung des Universums

    6. Temperaturentwicklung

    7. Kosmische Hintergrundstrahlung

    8. CMB kombiniert mit SN1a

    9. Strukturbildung

    10. Neutrinos

    11. Grand Unified Theories

    12.-13 Suche nach DM

    HEUTE

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 2

    Vorlesung 8

    Roter Faden:

    1. Powerspektrum der CMB

    2. Baryonic Acoustic Oscillations (BAO)

    3. Energieinhalt des Universums

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 3

    Akustische Peaks von WMAP

    Ort-Zeit

    Diagramm

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 4

    Kugelflächenfunktionen

    Jede Funktion kann in orthogonale

    Kugelflächenfkt. entwickelt werden. Große

    Werte von l beschreiben Korrelationen unter

    kleinen Winkel.

    l

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 5

    Lineweaver 1997

    peak

    trough

    Sky Maps Power Spectra

    We “see” the CMB sound

    as waves on the sky.

    Use special methods

    to measure the strength

    of each wavelength.

    Shorter wavelengths

    are smaller frequencies

    are higher pitches

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 6

    • Temperaturverteilung ist

    Funktion auf Sphäre:

    ΔT(θ,φ) bzw. ΔT(n) = ΔΘ(n)

    T T

    n=(sinθcosφ,sinθsinφ,cosθ)

    • Autokorrelationsfunktion:

    C(θ)=|n1-n2|

    =(4π)-1 Σ∞l=0 (2l+1)ClPl(cosθ)

    • Pl sind die Legendrepolynome:

    Pl(cosθ) = 2-l∙dl/d(cos θ)l(cos²θ-1)l.

    • Die Koeffizienten Cl bilden das Powerspektrum von ΔΘ(n).

    mit cosθ=n1∙n2

    Vom Bild zum Powerspektrum

    „Weißes Rauschen“:

    flaches Powerspektrum

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 7

    Temperaturschwankungen als Fkt. des Öffnungswinkels

    Θ 180/l Balloon exp.

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 8

    Das Leistungsspektrum (power spectrum)

    Ursachen für Temperatur-

    Schwankungen:

    Große Skalen:

    Gravitationspotentiale

    Kleine Skalen:

    Akustische Wellen

    l=1 nicht gezeigt, da

    sehr stark wegen

    Dipolterm durch

    Bewegung der Galaxie

    gegenüber CMB

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 9

    Temperaturanisotropie der CMB

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 10

    Position des ersten akustischen Peaks bestimmt

    Krümmung des Universums!

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 11

    Position des ersten Peaks

    Berechnung der Winkel, worunter man

    die maximale Temperaturschwankungen

    der Grundwelle beobachtet:

    Maximale Ausdehnung einer akust. Welle

    zum Zeitpunkt trec: cs * trec (1+z)

    Beobachtung nach t0 =13.8 109 yr.

    Öffnungswinkel θ = cs * trec * (1+z) / c*t0 Mit (1+z)= 3000/2.7 =1100 und

    trec = 3,8 105 yr und Schallgeschwindigkeit

    cs=c/3 für ein relativ. Plasma folgt:

    θ = 0.0175 = 10 (plus (kleine) ART Korrekt.)

    Beachte: cs2 ≡ dp/d = c2/3, da p= 1/3 c2

    Raum-Zeit x

    t Inflation

    Entkopplung

    max. T / T

    unter 10

    nλ/2=cstr

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 12

    Erste akustische Peak unter bei einem

    Öffnungswinkel von 0.8 Grad oder l=220

    bedeutet:

    das Universum ist flach

    oder die mittlere Dichte entspricht der

    kritischen Dichte von 2. 10-29 g/cm3 oder =1

    und

    Gesamtenergie (kin. + pot. Energie) ist Null!

    CMB zeigt: Un iversum ist flach

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 13

    Präzisere Berechnung des ersten Peaks

    Vor Entkopplung Universum teilweise strahlungsdominiert.

    Hier ist die Expansion t1/2 statt t2/3 in materiedominiertes Univ.

    Muss Abstände nach bewährtem Rezept berechnen:

    Erst in mitbewegten Koor. und dann x S(t)

    Abstand < trek: S(t) c d = S(t) c dt/S(t) = 2ctrek für S t1/2

    Abstand > trek: S(t) c d =S(t)c dt/S(t) = 3ctrek für S t2/3

    Winkel θ = 2 * cs * trec * (1+z) / 3*c*t0 = 0.7 Grad

    Auch nicht ganz korrekt, denn Univ. strahlungsdom. bis t=50000 a,

    nicht 380000 a. Richtige Antwort: Winkel θ = 0.8 Grad oder l=180/0.8=220

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 14

    http://wmap.gsfc.nasa.gov/resources/camb_tool/index.html

    WMAP analyzer tool

    http://wmap.gsfc.nasa.gov/resources/camb_tool/index.html

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 15

    Neueste WMAP Daten (2008)

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 16

    http://arxiv.org/PS_cache/arxiv/pdf/0803/0803.0732v2.pdf

    Neueste WMAP Daten (2008)

    Polarisation

    Temperatur

    Temperatur- und Polarisationsanisotropien um 90 Grad in Phase verschoben,

    weil Polarisation Fluss der Elektronen, also wenn x cos (t), dann v sin (t)

    Reionisation

    nach 2.108 a

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 17

    CMB Polarisation durch Thomson Streuung

    (elastische Photon-Electron Streuung)

    Prinzip: unpolarisiertes Photon unter 90 Grad gestreut, muss immer

    noch E-Feld Richtung haben, so eine Komponente verschwindet!

    Daher bei Isotropie keine Pol. , bei Dipol auch nicht, nur bei Quadr.

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 18

    Polarization entweder radial oder tangential um hot oder cold spots

    (proportional zum Fluss der Elektronen, also zeigt wie Plasma sich

    bewegte bei z=1100 and auf große Skalen wie Plasma in Galaxien

    Cluster sich relativ zum CMB bewegt)

    CMB Polarisation bei Quadrupole-Anisotropie

    http://gyudon.as.utexas.edu/~komatsu/presentation/wmap7_ias.pdf

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 19

    Entwicklung des Universums

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 20

    CMB polarisiert durch Streuung an Elektronen

    (Thomson Streuung)

    Kurz vor Entkoppelung:

    Streuung der CMB Photonen.

    Nachher nicht mehr, da mittlere

    freie Weglange zu groß.

    Lange vor der Entkopplung:

    Polarisation durch Mittelung

    über viele Stöße verloren.

    Nach Reionisation der Baryonen

    durch Sternentstehung wieder

    Streuung.

    Erwarte Polarisation also kurz

    nach dem akust. Peak (l = 300)

    und auf großen Abständen (l < 10) Instruktiv:http://background.uchicago.edu/~whu/polar/webversion/node1.htm

    l

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 21

    = x/S(t) = x(1+z)

    Raum-Zeit x

    t

    = t / S(t) = t (1+z)

    Conformal Space-Time

    (winkel-erhaltende Raum-Zeit)

    conformal=winkelerhaltend

    z.B. mercator Projektion

    x

    t

    t

    From Ned Wright homepage

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 22

    If it is not dark,

    it does not matter

    Woher kennt man diese Verteilung?

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 23

    Vergleich mit den SN 1a Daten

    SN1a empfindlich für

    Beschleunigung a, d.h.

    a - m (beachte:

    DM und DE unterschei-

    den sich im VZ der Grav.

    CMB empfindlich für

    totale Dichte d.h.

    tot = + m =1

    = (SM+ DM)

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 24

    Let's consider what happens to a point-

    like initial perturbation. In other words,

    we're going to take a little patch of space

    and make it a little denser. Of course, the

    universe has many such patchs, some

    overdense, some underdense. We're just

    going to focus on one. Because the

    fluctuations are so small, the effects of

    many regions just sum linearly.

    The relevant components of the universe

    are the dark matter, the gas (nuclei and

    electrons), the cosmic microwave

    background photons, and the cosmic

    background neutrinos.

    Akustische Baryon Oszillationen I: http://cmb.as.arizona.edu/~eisenste/acousticpeak/acoustic_physics.html

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 25

    Akustische Baryon Oszillationen II: http://cmb.as.arizona.edu/~eisenste/acousticpeak/acoustic_physics.html

    Now what happens?

    The neutrinos don't interact with anything and are too fast

    to be bound gravitationally, so they begin to stream away

    from the initial perturbation.

    The dark matter moves only in response to gravity and has

    no intrinsic motion (it's cold dark matter). So it sits still.

    The perturbation (now dominated by the photons and

    neutrinos) is overdense, so it attracts the surroundings,

    causing more dark matter to fall towards the center.

    The gas, however, is so hot at this time that it is ionized. In

    the resulting plasma, the cosmic microwave background

    photons are not able to propagate very far before they

    scatter off an electron. Effectively, the gas and photons are

    locked into a single fluid. The photons are so hot and

    numerous, that this combined fluid has an enormous

    pressure relative to its density. The initial overdensity is

    therefore also an initial overpressure. This pressure tries

    to equalize itself with the surroundings, but this simply

    results in an expanding spherical sound wave. This is just

    like a drum head pushing a sound wave into the air, but

    the speed of sound at this early time is 57% of the speed of

    light!

    The result is that the perturbation in

    the gas and photon is carried outward:

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 26

    Akustische Baryon Oszillationen III: http://cmb.as.arizona.edu/~eisenste/acousticpeak/acoustic_physics.html

    As time goes on, the spherical shell of gas

    and photons continues to expand. The

    neutrinos spread out. The dark matter

    collects in the overall density perturbation,

    which is now considerably bigger because

    the photons and neutrinos have left the

    center. Hence, the peak in the dark matter

    remains centrally concentrated but with an

    increasing width. This is generating the

    familiar turnover in the cold dark matter

    power spectrum.

    Where is the extra dark matter at large

    radius coming from? The gravitational

    forces are attracting the background

    material in that region, causing it to contract

    a bit and become overdense relative to the

    background further away

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 27

    Akustische Baryon Oszillationen IV: http://cmb.as.arizona.edu/~eisenste/acousticpeak/acoustic_physics.html

    The expanding universe is cooling.

    Around 400,000 years, the

    temperature is low enough that the

    electrons and nuclei begin to combine

    into neutral atoms. The photons do

    not scatter efficiently off of neutral

    atoms, so the photons begin to slip

    past the gas particles. This is known

    as Silk damping (ApJ, 151, 459, 1968).

    The sound speed begins to drop

    because of the reduced coupling

    between the photons and gas and

    because the cooler photons are no

    longer very heavy compared to the

    gas. Hence, the pressure wave slows

    down.

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 28

    Akustische Baryon Oszillationen V: http://cmb.as.arizona.edu/~eisenste/acousticpeak/acoustic_physics.html

    This continues until the photons have

    completely leaked out of the gas

    perturbation. The photon perturbation

    begins to smooth itself out at the speed

    of light (just like the neutrinos did).

    The photons travel (mostly)

    unimpeded until the present-day,

    where we can record them as the

    microwave background (see below).

    At this point, the sound speed in the

    gas has dropped to much less than the

    speed of light, so the pressure wave

    stalls.

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 29

    Akustische Baryon Oszillationen VI: http://cmb.as.arizona.edu/~eisenste/acousticpeak/acoustic_physics.html

    We are left with a dark matter

    perturbation around the original

    center and a gas perturbation in a shell

    about 150 Mpc (500 million light-

    years) in radius.

    As time goes on, however, these two

    species gravitationally attract each

    other. The perturbations begin to mix

    together. More precisely, both

    perturbations are growing quickly in

    response to the combined gravitational

    forces of both the dark matter and the

    gas. At late times, the initial

    differences are small compared to the

    later growth.

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 30

    Akustische Baryon Oszillationen VII: http://cmb.as.arizona.edu/~eisenste/acousticpeak/acoustic_physics.html

    Eventually, the two look quite

    similar. The spherical shell of the

    gas perturbation has imprinted

    itself in the dark matter. This is

    known as the acoustic peak.

    The acoustic peak decreases in

    contrast as the gas come into lock-

    step with the dark matter simply

    because the dark matter, which has

    no peak initially, outweighs the gas

    5 to 1.

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 31

    Akustische Baryon Oszillationen VIII: http://cmb.as.arizona.edu/~eisenste/acousticpeak/acoustic_physics.html

    At late times, galaxies form in the

    regions that are overdense in gas and

    dark matter. For the most part, this is

    driven by where the initial

    overdensities were, since we see that the

    dark matter has clustered heavily

    around these initial locations. However,

    there is a 1% enhancement in the

    regions 150 Mpc away from these

    initial overdensities. Hence, there

    should be an small excess of galaxies

    150 Mpc away from other galaxies, as

    opposed to 120 or 180 Mpc. We can see

    this as a single acoustic peak in the

    correlation function of galaxies.

    Alternatively, if one is working with the

    power spectrum statistic, then one sees

    the effect as a series of acoustic

    oscillations.

    Before we have been plotting the mass profile

    (density times radius squared). The density

    profile is much steeper, so that the peak at 150

    Mpc is much less than 1% of the density near

    the center.

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 32

    One little telltale bump !!

    A small excess in correlation at 150 Mpc.!

    SDSS survey

    (astro-ph/0501171)

    150 Mpc.

    (Einsentein et al. 2005)

    1 2( ) ( ) ( )r r r

    150 Mpc =2cs tr (1+z)=akustischer Horizont

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 33

    The same CMB oscillations at

    low redshifts !!!

    SDSS survey

    (astro-ph/0501171)

    150 Mpc.

    (Eisentein et al. 2005)

    105 h-1 ¼ 150

    Akustische Baryonosz. in Korrelationsfkt. der

    Dichteschwankungen der Materie!

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 34

    http://nedwww.ipac.caltech.edu/level5/March08/Frieman/Frieman4.html

    Combined results

    http://arxiv.org/PS_cache/arxiv/pdf/0804/0804.4142v1.pdf

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 35

    Zum Mitnehmen

    Die CMB gibt ein Bild des frühen Universums 380.000 yr nach dem Urknall und zeigt

    die Dichteschwankungen T/T, woraus später die Galaxien entstehen.

    Die CMB zeigt dass

    1. das Univ. am Anfang heiß war, weil akustische Peaks, entstanden

    durch akustische stehende Wellen in einem heißen Plasma, entdeckt wurden

    2. die Temperatur der Strahlung im Universum 2.7 K ist wie erwartet bei einem

    EXPANDIERENDEN Univ. mit Entkopplung der heißen Strahlung und Materie

    bei einer Temp. von 3000 K oder z=1100 (T 1+z !)

    3. das Univ. FLACH ist, weil die Photonen sich seit der letzten Streuung

    zum Zeitpunkt der Entkopplung (LSS = last scattering surface) auf gerade

    Linien bewegt haben (in comoving coor.)

    4) BAO wichtig, weil Sie unabhängig von der akustischen Horizont in der CMB ein

    zweiter wohl definierter Maßstab (akustischer Horizont der Materie) bestimmt,

    dessen Vergrößerung heute gemessen werden kann. Dies bestätigt die

    Energieverteilung des Univ. unabh. von der Frage ob SN1a Standardkerzen sind.

    5) Polarisation der CMB bestätigt Natur der Dichtefluktuationen zum Zeitpunkt der

    Entkopplung und bestimmt Zeitpunkt der Sternbildung (Ionisation->Polarisation)

    Die schnelle Sternbildung kann nur mit Potentialtöpfen der DM zum Zeitpunkt der

    Entkopplung erklärt werden. (die neutrale Kerne fallen da hinein).

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 36

    If it is not dark,

    it does not matter

    Zum Mitnehmen

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 37

    Zusatzfolien mit Text der Nobelpreisankündigungen

    „just for fun“, kein Prüfungsstoff.

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 38

    The Universe is approximately about 13.7 billion years old, according to the

    standard cosmological Big Bang model. At this time, it was a state of high

    uniformity, was extremely hot and dense was filled with elementary particles

    and was expanding very rapidly. About 380,000 years after the Big Bang, the

    energy of the photons had decreased and was not sufficient to ionise hydrogen

    atoms. Thereafter the photons “decoupled” from the other particles and could

    move through the Universe essentially unimpeded. The Universe has expanded

    and cooled ever since, leaving behind a remnant of its hot past, the Cosmic

    Microwave Background radiation (CMB). We observe this today as a 2.7 K

    thermal blackbody radiation filling the entire Universe. Observations of the

    CMB give a unique and detailed information about the early Universe, thereby

    promoting cosmology to a precision science. Indeed, as will be discussed in

    more detail below, the CMB is probably the best recorded blackbody spectrum

    that exists. Removing a dipole anisotropy, most probably due our motion

    through the Universe, the CMB is isotropic to about one part in 100,000. The

    2006 Nobel Prize in physics highlights detailed observations of the CMB

    performed with the COBE (COsmic Background Explorer) satellite.

    Cosmology and the Cosmic Microwave Background

    From Nobel prize 2006 announcement

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 39

    The discovery of the cosmic microwave background radiation has an

    unusual and interesting history. The basic theories as well as the necessary

    experimental techniques were available long before the experimental

    discovery in 1964. The theory of an expanding Universe was first given by

    Friedmann (1922) and Lemaître (1927). An excellent account is given by

    Nobel laureate Steven Weinberg (1993).

    Around 1960, a few years before the discovery, two scenarios for the

    Universe were discussed. Was it expanding according to the Big Bang

    model, or was it in a steady state? Both models had their supporters and

    among the scientists advocating the latter were Hannes Alfvén (Nobel prize

    in physics 1970), Fred Hoyle and Dennis Sciama. If the Big Bang model

    was the correct one, an imprint of the radiation dominated early Universe

    must still exist, and several groups were looking for it. This radiation must

    be thermal, i.e. of blackbody form, and isotropic.

    Early work

    From Nobel prize 2006 announcement

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 40

    The discovery of the cosmic microwave background by Penzias and Wilson in 1964

    (Penzias and Wilson 1965, Penzias 1979, Wilson 1979, Dicke et al. 1965) came as a

    complete surprise to them while they were trying to understand the source of

    unexpected noise in their radio-receiver (they shared the 1978 Nobel prize in

    physics for the discovery). The radiation produced unexpected noise in their radio

    receivers. Some 16 years earlier Alpher, Gamow and Herman (Alpher and Herman

    1949, Gamow 1946), had predicted that there should be a relic radiation field

    penetrating the Universe. It had been shown already in 1934 by Tolman (Tolman

    1934) that the cooling blackbody radiation in an expanding Universe retains its

    blackbody form. It seems that neither Alpher, Gamow nor Herman succeeded in

    convincing experimentalists to use the characteristic blackbody form of the

    radiation to find it. In 1964, however, Doroshkevich and Novikov (Doroshkevich

    and Novikov 1964) published an article where they explicitly suggested a search for

    the radiation focusing on its blackbody characteristics. One can note that some

    measurements as early as 1940 had found that a radiation field was necessary to

    explain energy level transitions in interstellar molecules (McKellar 1941).

    Following the 1964 discovery of the CMB, many, but not all, of the steady state

    proponents gave up, accepting the hot Big Bang model. The early theoretical work

    is discussed by Alpher, Herman and Gamow 1967, Penzias 1979, Wilkinson and

    Peebles 1983, Weinberg 1993, and Herman 1997.

    First observations of CMB

    CN=Cyan

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 41

    Following the 1964 discovery, several independent measurements of the

    radiation were made by Wilkinson and others, using mostly balloon-borne,

    rocket-borne or ground based instruments. The intensity of the radiation has

    its maximum for a wavelength of about 2 mm where the absorption in the

    atmosphere is strong. Although most results gave support to the blackbody

    form, few measurements were available on the high frequency (low

    wavelength) side of the peak. Some measurements gave results that showed

    significant deviations from the blackbody form (Matsumoto et al. 1988).

    The CMB was expected to be largely isotropic. However, in order to explain

    the large scale structures in the form of galaxies and clusters of galaxies

    observed today, small anisotropies should exist. Gravitation can make small

    density fluctuations that are present in the early Universe grow and make

    galaxy formation possible. A very important and detailed general relativistic

    calculation by Sachs and Wolfe showed how three-dimensional density

    fluctuations can give rise to two-dimensional large angle (> 1°) temperature

    anisotropies in the cosmic microwave background radiation (Sachs and

    Wolfe 1967).

    Further observations of CMB

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 42

    Because the earth moves relative to the CMB, a dipole temperature

    anisotropy of the level of ΔT/T = 10-3 is expected. This was observed in the

    1970’s (Conklin 1969, Henry 1971, Corey and Wilkinson 1976 and Smoot,

    Gorenstein and Muller 1977). During the 1970-tis the anisotropies were

    expected to be of the order of 10-2 – 10-4, but were not observed

    experimentally. When dark matter was taken into account in the 1980-ties,

    the predicted level of the fluctuations was lowered to about 10-5, thereby

    posing a great experimental challenge.

    Dipol Anisotropy

    Explanation: two effects compensate the temperature anisotropies:

    DM dominates the gravitational potential after str

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 43

    Because of e.g. atmospheric absorption, it was long realized that

    measurements of the high frequency part of the CMB spectrum

    (wavelengths shorter than about 1 mm) should be performed from

    space. A satellite instrument also gives full sky coverage and a long

    observation time. The latter point is important for reducing systematic

    errors in the radiation measurements. A detailed account of

    measurements of the CMB is given in a review by Weiss (1980).

    The COBE story begins in 1974 when NASA made an announcement of opportunity

    for small experiments in astronomy. Following lengthy discussions with NASA

    Headquarters the COBE project was born and finally, on 18 November 1989, the

    COBE satellite was successfully launched into orbit. More than 1,000 scientists,

    engineers and administrators were involved in the mission. COBE carried three

    instruments covering the wavelength range 1 μm to 1 cm to measure the anisotropy

    and spectrum of the CMB as well as the diffuse infrared background radiation:

    DIRBE (Diffuse InfraRed Background Experiment), DMR (Differential Microwave

    Radiometer) and FIRAS (Far InfraRed Absolute Spectrophotometer). COBE’s

    mission was to measure the CMB over the entire sky, which was possible with the

    chosen satellite orbit. All previous measurements from ground were done with limited

    sky coverage. John Mather was the COBE Principal Investigator and the project

    leader from the start. He was also responsible for the FIRAS instrument. George

    Smoot was the DMR principal investigator and Mike Hauser was the DIRBE principal

    investigator.

    The COBE mission

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 44

    For DMR the objective was to search for anisotropies at three

    wavelengths, 3 mm, 6 mm, and 10 mm in the CMB with an

    angular resolution of about 7°. The anisotropies postulated to

    explain the large scale structures in the Universe should be

    present between regions covering large angles. For FIRAS

    the objective was to measure the spectral distribution of the

    CMB in the range 0.1 – 10 mm and compare it with the

    blackbody form expected in the Big Bang model, which is

    different from, e.g., the forms expected from starlight or

    bremsstrahlung. For DIRBE, the objective was to measure

    the infrared background radiation. The mission, spacecraft

    and instruments are described in detail by Boggess et al.

    1992. Figures 1 and 2 show the COBE orbit and the satellite,

    respectively.

    The COBE mission

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 45

    COBE was a success. All instruments worked very

    well and the results, in particular those from DMR

    and FIRAS, contributed significantly to make

    cosmology a precision science. Predictions of the Big

    Bang model were confirmed: temperature

    fluctuations of the order of 10-5 were found and the

    background radiation with a temperature of 2.725 K

    followed very precisely a blackbody spectrum.

    DIRBE made important observations of the infrared

    background. The announcement of the discovery of

    the anisotropies was met with great enthusiasm

    worldwide.

    The COBE success

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 46

    The DMR instrument (Smoot et al. 1990) measured temperature

    fluctuations of the order of 10-5 for three CMB frequencies, 90, 53 and

    31.5 GHz (wavelengths 3.3, 5.7 and 9.5 mm), chosen near the CMB

    intensity maximum and where the galactic background was low. The

    angular resolution was about 7°. After a careful elimination of

    instrumental background, the data showed a background contribution

    from the Milky Way, the known dipole amplitude ΔT/T = 10-3 probably

    caused by the Earth’s motion in the CMB, and a significant long sought

    after quadrupole amplitude, predicted in 1965 by Sachs and Wolfe. The

    first results were published in 1992.The data showed scale invariance for

    large angles, in agreement with predictions from inflation models.

    Figure 5 shows the measured temperature fluctuations in galactic coordinates, a figure

    that has appeared in slightly different forms in many journals. The RMS cosmic

    quadrupole amplitude was estimated at 13 ± 4 μK (ΔT/T = 5×10-6) with a systematic

    error of at most 3 μK (Smoot et al. 1992). The DMR anisotropies were compared and

    found to agree with models of structure formation by Wright et al. 1992. The full 4 year

    DMR observations were published in 1996 (see Bennett et al. 1996). COBE’s results

    were soon confirmed by a number of balloon-borne experiments, and, more recently, by

    the 1° resolution WMAP (Wilkinson Microwave Anisotropy Probe) satellite, launched

    in 2001 (Bennett et al. 2003).

    CMB Anisotropies

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 47

    The 1964 discovery of the cosmic microwave background had a large impact

    on cosmology. The COBE results of 1992, giving strong support to the Big

    Bang model, gave a much more detailed view, and cosmology turned into a

    precision science. New ambitious experiments were started and the rate of

    publishing papers increased by an order of magnitude.

    Our understanding of the evolution of the Universe rests on a number of observations,

    including (before COBE) the darkness of the night sky, the dominance of hydrogen and

    helium over heavier elements, the Hubble expansion and the existence of the CMB.

    COBE’s observation of the blackbody form of the CMB and the associated small

    temperature fluctuations gave very strong support to the Big Bang model in proving

    the cosmological origin of the CMB and finding the primordial seeds of the large

    structures observed today.

    However, while the basic notion of an expanding Universe is well established,

    fundamental questions remain, especially about very early times, where a nearly

    exponential expansion, inflation, is proposed. This elegantly explains many

    cosmological questions. However, there are other competing theories. Inflation may

    have generated gravitational waves that in some cases could be detected indirectly by

    measuring the CMB polarization. Figure 8 shows the different stages in the evolution

    of the Universe according to the standard cosmological model. The first stages after the

    Big Bang are still speculations.

    Outlook

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 48

    The young Universe was fantastically bright. Why? Because everywhere it

    was hot, and hot things glow brightly. Before we learned why this was:

    collisions between charged particles create photons of light. As long as the

    particles and photons can thoroughly interact then a thermal spectrum is

    produced: a broad range with a peak.

    The thermal spectrum’s shape depends only on temperature: Hotter objects

    appear bluer: the peak shifts to shorter wavelengths, with: pk = 0.0029/TK

    m = 2.9106/T nm. At 10,000K we have peak = 290 nm (blue), while at

    3000K we have peak = 1000 nm (deep orange/red).

    Let’s now follow through the color of the Universe during its first million

    years. As the Universe cools, the thermal spectrum shifts from blue to red,

    spending ~80,000 years in each rainbow color.

    At 50 kyr, the sky is blue! At 120 kyr it’s green; at 400 kyr it’s orange; and

    by 1 Myr it’s crimson. This is a wonderful quality of the young Universe: it

    paints its sky with a human palette.

    Quantitatively: since peak ~ 3106/T nm, and T ~ 3/S K, then peak ~ 106 /

    S nm. Notice that today, S = 1 and so peak = 106 nm = 1 mm, which is, of

    course, the peak of the CMB microwave spectrum.

    The colour of the universe

  • Wim de Boer, Karlsruhe

    Kosmologie VL, 13.12.2012 49

    Hotter objects appear brighter. There are two reasons for this:

    More violent particle collisions make more energetic photons. Converting pk ~

    0.003/T m to the equivalent energy units, it turns out that in a thermal spectrum,

    the average photon energy is ~ kT. So, for systems in thermal equilibrium, the

    mean energy per particle or per photon is ~kT. Faster particles collide more

    frequently, so make more photons. In fact the number density of photons, nph

    T3. Combining these, we find that the intensity of thermal radiation increases

    dramatically with temperature Itot = 2.210-7 T4 Watt /m2 inside a gas at

    temperature T.

    At high temperatures, thermal radiation has awesome power – the multitude of particle

    collisions is incredibly efficient at creating photons. To help feel this, consider the light

    falling on you from a noontime sun – 1400 Watt/m2 – enough to feel sunburned quite

    quickly. Let’s write this as Isun.

    Float in outer space, exposed only to the CMB, and you experience a radiation

    field of I3K = 2.210-72.74 = 10 W/m2 = 10-8 Isun – not much! Here on Earth at

    300K we have I300K ~ 1.8 kW/m2 (fortunately, our body temperature is 309K so

    you radiate 2.0 kW/m2, and don’t quickly boil!). A blast furnace at 1500 C

    (~1800K) has I1800K = 2.3 MW/m2 = 1600 Isun (you boil away in ~1 minute).

    At the time of the CMB (380 kyr), the radiation intensity was I3000K = 17 MW/m2

    = 12,000 Isun – you evaporate in 10 seconds.

    In the Sun’s atmosphere, we have I5800K = 250 MW/m2 = 210,000 Isun. That’s a

    major city’s power usage, falling on each square meter.

    Radiation in the Sun’s 14 million K core has: I = 81021 W/m2 ~ 1019 Isun (you

    boil away in much less than a nano-second).

    Light Intensity