Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Why cosmology should care about the Milky Way

Daniel
September 28, 2017

Why cosmology should care about the Milky Way

JPL Postdoc Seminar Series talk on cosmological foregrounds

Daniel

September 28, 2017
Tweet

More Decks by Daniel

Other Decks in Science

Transcript

  1. Why cosmology should
    care about the Milky Way
    in collaboration with O. Doré,
    B. Hensley, P. Bull, G. Lagache,
    P. Serra
    Daniel Lenz
    JPL Postdoc Seminar Series
    Sept 28, 2017
    © 2017 California Institute of Technology. Government sponsorship acknowledged.

    View full-size slide

  2. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Why should cosmologists care
    about the Milky Way?
    2
    'Galactic'
    astronomers…
    ❖ … want to
    understand how
    galaxies are formed/
    evolve/merge etc.
    ❖ … observe and
    model our Galaxy in
    different wavelengths

    View full-size slide

  3. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    3
    Why should cosmologists care
    about the Milky Way?
    'Cosmologists' …
    ❖ … want to
    understand the
    history, evolution
    and fate of the entire
    Universe
    ❖ …study millions of
    galaxies and their
    statistical properties Hubble Ultra Deep Field

    View full-size slide

  4. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    3
    Why should cosmologists care
    about the Milky Way?
    'Cosmologists' …
    ❖ … want to
    understand the
    history, evolution
    and fate of the entire
    Universe
    ❖ …study millions of
    galaxies and their
    statistical properties Hubble Ultra Deep Field
    Warm dark matter
    Cold dark matter
    Bromm+ (2009)

    View full-size slide

  5. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    4
    Why should cosmologists care
    about the Milky Way?
    'Cosmologists' …
    ❖ … want to understand
    the history, evolution
    and fate of the entire
    Universe
    ❖ …study cosmological
    backgrounds
    Cosmic Microwave Background (Planck)

    View full-size slide

  6. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    5
    Why should cosmologists care
    about the Milky Way?
    NASA/WMAP

    View full-size slide

  7. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    6
    Why should cosmologists care
    about the Milky Way?

    View full-size slide

  8. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Cosmology and the Milky Way: Synergies
    7

    View full-size slide

  9. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Cosmology and the Milky Way: Synergies
    ❖ Our Galaxy is an inconvenient foreground for Cosmology
    7

    View full-size slide

  10. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Cosmology and the Milky Way: Synergies
    ❖ Our Galaxy is an inconvenient foreground for Cosmology
    ❖ Cosmology needs a model of the Milky Way to remove it
    7

    View full-size slide

  11. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Cosmology and the Milky Way: Synergies
    ❖ Our Galaxy is an inconvenient foreground for Cosmology
    ❖ Cosmology needs a model of the Milky Way to remove it
    ❖ Galactic astronomers benefit from cosmology missions,
    giving them fantastic data to work with
    7

    View full-size slide

  12. Which foregrounds do we care about?
    "(…) the name of the game is component separation, not noise
    reduction"
    H.K. Eriksen, 'Advances in Theoretical Cosmology in Light of Data 2017'
    ❖ Extinction for cosmological galaxy surveys
    ❖ Cosmic infrared background measurements
    ❖ De-lensing of CMB data for primordial gravitational waves
    Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    8

    View full-size slide

  13. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    E(B-V)
    ❖ E(B-V) = Extinction in B band - Extinction in V band
    ❖ More dust => larger E(B-V)
    ❖ E(B-V) maps essential for correcting observations for
    Galactic reddening
    10

    View full-size slide

  14. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Mapping E(B-V): Direct approach
    11
    ❖ Find many sources with
    known spectrum (e.g.
    stars, passive galaxies)
    ❖ Measure spectra, attribute
    differences to dust
    ❖ E.g. Schlafly+ (2014),
    Green+ (2015) used 500
    million stars from Pan-
    STARRS to measure
    reddening to 4.5 kpc

    View full-size slide

  15. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Mapping E(B-V): Direct approach
    11
    ❖ Find many sources with
    known spectrum (e.g.
    stars, passive galaxies)
    ❖ Measure spectra, attribute
    differences to dust
    ❖ E.g. Schlafly+ (2014),
    Green+ (2015) used 500
    million stars from Pan-
    STARRS to measure
    reddening to 4.5 kpc
    ❖ Direct measurements are
    hard!
    ❖ Photometric/
    spectroscopic errors
    ❖ Ensuring sources lie
    behind full dust column
    ❖ Ensuring adequate
    number of sources have
    been measured

    View full-size slide

  16. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Mapping E(B-V): Direct approach
    11
    ❖ Find many sources with
    known spectrum (e.g.
    stars, passive galaxies)
    ❖ Measure spectra, attribute
    differences to dust
    ❖ E.g. Schlafly+ (2014),
    Green+ (2015) used 500
    million stars from Pan-
    STARRS to measure
    reddening to 4.5 kpc
    ❖ Direct measurements are
    hard!
    ❖ Photometric/
    spectroscopic errors
    ❖ Ensuring sources lie
    behind full dust column
    ❖ Ensuring adequate
    number of sources have
    been measured
    Green+ (2015)

    View full-size slide

  17. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Dust emission as measure of E(B-V)
    ❖ E(B-V) is proportional to the
    dust column, so can convert
    dust column tracer to E(B-V)
    ❖ SFD used dust emission from
    IRAS to derive a calibration
    factor from FIR emission to
    E(B-V)
    ❖ Full-sky, high sensitivity
    measurements
    -2 -0.3
    log10
    (E(B V )SFD
    [mag])
    Reddening map of Schlegel, Finkbeiner,
    and Davis (1998)
    12

    View full-size slide

  18. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    ❖ Requires a temperature
    correction to go from dust
    emission to a dust column
    density
    ❖ FIR emission may have
    contributions from zodiacal
    light and unresolved galaxies
    -2 -0.3
    log10
    (E(B V )SFD
    [mag])
    Reddening map of Schlegel, Finkbeiner,
    and Davis (1998)
    13
    Dust emission as measure of E(B-V)

    View full-size slide

  19. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    ❖ Requires a temperature
    correction to go from dust
    emission to a dust column
    density
    ❖ FIR emission may have
    contributions from zodiacal
    light and unresolved galaxies
    -2 -0.3
    log10
    (E(B V )SFD
    [mag])
    Reddening map of Schlegel, Finkbeiner,
    and Davis (1998)
    13
    Dust emission as measure of E(B-V)

    View full-size slide

  20. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    HI emission as basis for E(B-V)
    ❖ Neutral atomic hydrogen (HI), most abundant element
    ❖ Gas and dust are well-coupled in the interstellar medium
    ❖ Resulting maps free from errors due to dust temperature,
    zodi, and extragalactic emission
    ❖ Limited by non-HI gas along the line of sight
    14

    View full-size slide

  21. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Correlation of dust and gas
    HI Dust
    15

    View full-size slide

  22. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Galactic HI Surveys
    16
    Credit: S. Janowiecki
    Southern hemisphere:
    Galactic All-Sky
    Survey (GASS)
    McClure-Griffiths+
    (2009)

    Kalberla+ (2010, 2015)
    D. McClenaghan

    View full-size slide

  23. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Galactic HI Surveys
    16
    B. Winkel
    Northern hemisphere:
    Effelsberg-Bonn HI
    Survey (EBHIS)
    Kerp+ (2011)

    Winkel, DL+ (2010, 2016)

    View full-size slide

  24. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    HI4PI Survey
    ❖ Merges data from Effelsberg and Parkes
    ❖ New state-of-the-art survey
    ❖ Higher sensitivity & resolution, fewer systematics, full sampling
    20
    21
    22
    log(NHI
    [cm 2])
    180
    135 90
    45
    0
    315
    270
    225 180
    60
    30
    0
    30
    60
    HI4PI collaboration

    (2017)
    17

    View full-size slide

  25. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    HI emission as basis for E(B-V)
    ❖ Use HI4PI column density as new measure of dust
    reddening
    ❖ How well does it actually fit other reddening tracers?
    19

    View full-size slide

  26. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    10 1
    100
    (B V ) [mag]
    0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
    NHI [cm 2] ⇥1020
    0.00
    0.01
    0.02
    0.03
    0.04
    0.05
    0.06
    0.07
    0.08
    E(B V ) [mag]
    E(B V ) [mag] = 1.216+0.009
    0.009
    ⇥ NHI [1022 cm 2] + 0.015+0.0002
    0.0002
    [mag] = 0.02406+0.00006
    0.00006
    102
    103
    104
    # data points
    0
    10
    20
    30
    40
    50
    60
    70
    80
    # data points
    10 1
    100
    (B V ) [mag]
    0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
    NHI [cm 2] ⇥1020
    0.00
    0.01
    0.02
    0.03
    0.04
    0.05
    0.06
    0.07
    0.08
    E(B V ) [mag]
    E(B V ) [mag] = 1.113+0.002
    0.002
    ⇥ NHI [1022 cm 2] + 0.000+0.0001
    0.0001
    [mag] = 0.00570+0.00001
    0.00001
    102
    103
    104
    # data points
    0
    100
    200
    300
    400
    # data points
    The E(B-V)/NHI ratio
    Pan-STARRS E(B-V), Schlafly+ (2014) SFD E(B-V)
    Star-based Dust-based
    20
    DL, Hensley, Doré (2017)

    View full-size slide

  27. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    10 1
    100
    (B V ) [mag]
    0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
    NHI [cm 2] ⇥1020
    0.00
    0.01
    0.02
    0.03
    0.04
    0.05
    0.06
    0.07
    0.08
    E(B V ) [mag]
    E(B V ) [mag] = 1.216+0.009
    0.009
    ⇥ NHI [1022 cm 2] + 0.015+0.0002
    0.0002
    [mag] = 0.02406+0.00006
    0.00006
    102
    103
    104
    # data points
    0
    10
    20
    30
    40
    50
    60
    70
    80
    # data points
    10 1
    100
    (B V ) [mag]
    0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
    NHI [cm 2] ⇥1020
    0.00
    0.01
    0.02
    0.03
    0.04
    0.05
    0.06
    0.07
    0.08
    E(B V ) [mag]
    E(B V ) [mag] = 1.113+0.002
    0.002
    ⇥ NHI [1022 cm 2] + 0.000+0.0001
    0.0001
    [mag] = 0.00570+0.00001
    0.00001
    102
    103
    104
    # data points
    0
    100
    200
    300
    400
    # data points
    The E(B-V)/NHI ratio
    Pan-STARRS E(B-V), Schlafly+ (2014) SFD E(B-V)
    Star-based Dust-based
    20
    DL, Hensley, Doré (2017)

    View full-size slide

  28. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    HI-based reddening model
    ❖ For the full sky, allow
    different E(B-V)/NHI
    ratios for different
    velocities
    ❖ High-velocity gas has
    less dust, as expected
    21
    102 101 0 101 102
    vLSR [km s 1]
    0.000
    0.005
    0.010
    0.015
    0.020
    0.025
    ↵v = E(B V )/Nv
    HI
    [mag/1020 cm 2]
    ↵v
    DL, Hensley, Doré (2017)

    View full-size slide

  29. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    130 112 95
    l [deg]
    42
    52
    62
    b [deg]
    SFD Modelsimple
    130 112 95
    l [deg]
    SFD Modeltophat
    0.00
    0.02
    0.04
    0.016
    0.008
    0.000
    0.008
    0.016
    E(B V ) [mag]
    102
    103
    104
    105
    # data points
    HI-based reddening model
    ❖ Black: HI high-velocity clouds
    ❖ Color: Reddening residuals
    22
    Full HI column density Only local HI (|v| < 90 km/s)
    DL, Hensley, Doré (2017)

    View full-size slide

  30. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    0 0.05
    E(B V )model
    [mag]
    The E(B-V) map
    40% sky coverage, 16.1’ resolution
    23
    DL, Hensley, Doré (2017)

    View full-size slide

  31. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Dust systematics
    ❖ Peek & Graves (2010) used
    SDSS passively evolving
    galaxies as "standard crayons"
    ❖ Correction to the SFD map at
    4.5 deg
    24

    View full-size slide

  32. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Dust systematics
    Based on extragalactic sources Based on galactic HI
    25

    View full-size slide

  33. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    When and why to use this extinction map
    ❖ New HI-based extinction map
    ❖ In line with independent corrections, but much higher
    resolution and better sky coverage
    ❖ Yahata+ (2007) find correlation of SFD with large-scale
    structure
    ❖ For high latitudes, our map overcomes many of the SFD
    problems and is much more sensitive than stellar data-
    based E(B-V) maps
    26

    View full-size slide

  34. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    The Cosmic Infrared Background
    What is the CIB?
    27

    View full-size slide

  35. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    The Cosmic Infrared Background
    What is the CIB?
    27

    View full-size slide

  36. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    The Cosmic Infrared Background
    ❖ Unresolved
    background radiation
    ❖ Made up from dust
    in galaxies at z=1-3
    Lagache+ (2002)
    28

    View full-size slide

  37. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    The Cosmic Infrared Background
    ❖ Unresolved
    background radiation
    ❖ Made up from dust
    in galaxies at z=1-3
    Lagache+ (2002)
    28

    View full-size slide

  38. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    The CIB as cosmological probe
    … of star formation history
    Planck collaboration (2013 XXX)
    ❖ Strong constraints on
    SFH up to z=2.5
    ❖ Probe dust temperature
    across cosmic times
    ❖ Understand star
    formation in DM halos
    29

    View full-size slide

  39. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    ❖ CMB lensing and CIB match great
    in z and MHalo
    ❖ Ideal probe of relation between
    dark and luminous matter
    … of large scale structure to
    cross-correlate with lensing
    Planck collaboration (2014 XVIII)
    The CIB as cosmological probe
    30

    View full-size slide

  40. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    ❖ CMB lensing and CIB match great
    in z and MHalo
    ❖ Ideal probe of relation between
    dark and luminous matter
    … of large scale structure to
    cross-correlate with lensing
    Planck collaboration (2014 XVIII)
    The CIB as cosmological probe
    30

    View full-size slide

  41. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    How to obtain CIB maps?
    ❖ Galactic thermal dust and CIB dust dominate on large scales at
    ~200 to 1000 GHz
    ❖ How to disentangle them?
    31

    View full-size slide

  42. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    How to obtain CIB maps?
    A. Fit different frequency channels with modified blackbody spectra
    B. Utilize the different angular power spectra of these components
    C. Use template maps of Galactic dust (e.g. HI-based)
    ❖ Galactic thermal dust and CIB dust dominate on large scales at
    ~200 to 1000 GHz
    ❖ How to disentangle them?
    31

    View full-size slide

  43. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Current CIB maps: Planck (2013 XXX)
    ❖ Limited sky coverage, hard to access large scales
    32

    View full-size slide

  44. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Modeling dust foregrounds

    • Velocity separation difficult for
    complex structures and large scales
    Radial Velocity
    HVC
    IVC
    LVC
    I = ✏HVC NHVC + ✏IVC NIVC + ✏LVC NLVC
    33

    View full-size slide

  45. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Modeling dust foregrounds
    • Generalised linear model (GLM)

    Radial Velocity
    I =
    X
    i
    ✏iTi
    B
    34

    View full-size slide

  46. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Modeling dust foregrounds
    • Generalised linear model (GLM)

    • Regularised:

    • Accounts for all features along
    line of sight
    I =
    X
    i
    ✏iTi
    B
    Radial Velocity
    |
    Datai Modeli
    |2
    +
    ↵ · |✏i
    |
    35

    View full-size slide

  47. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    CIB: Galactic poles
    Total FIR intensity
    36

    View full-size slide

  48. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    CIB: Galactic poles
    CIB
    36

    View full-size slide

  49. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    CIB: Challenges
    37

    View full-size slide

  50. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    CIB: Challenges
    ❖ How to verify, especially on large scales?
    ❖ Simulations and re-sampling
    37

    View full-size slide

  51. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    CIB: Challenges
    ❖ How to verify, especially on large scales?
    ❖ Simulations and re-sampling
    ❖ How to avoid fine-tuning of the component separation?
    37

    View full-size slide

  52. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    CIB: Challenges
    ❖ How to verify, especially on large scales?
    ❖ Simulations and re-sampling
    ❖ How to avoid fine-tuning of the component separation?
    ❖ How to extend this larger areas, probing non-HI gas?
    37

    View full-size slide

  53. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    CIB: Challenges
    ❖ How to verify, especially on large scales?
    ❖ Simulations and re-sampling
    ❖ How to avoid fine-tuning of the component separation?
    ❖ How to extend this larger areas, probing non-HI gas?
    ❖ How to jointly use frequency information and angular
    power spectra?
    37

    View full-size slide

  54. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    De-lensing the cosmic microwave
    background
    38

    View full-size slide

  55. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    De-lensing the CMB
    39
    ❖ Large-scale structure lenses the CMB, smoothes the power spectrum

    View full-size slide

  56. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    CMB de-lensing: TT

    ❖ Ideally: Construct lensing potential from CMB itself
    ❖ Too noisy, but will be relevant for CMB-S3 and CMB-S4
    ❖ Currently: Use external tracer of lensing potential such as the CIB
    42
    Te↵(
    x
    ) =
    Ttrue(
    x
    + r (
    x
    ))

    View full-size slide

  57. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    CMB de-lensing: TT
    43
    ❖ Using large-scale
    CIB maps to de-lens
    Planck CMB TT
    ❖ Characteristic
    sharpening of the
    peaks detected at
    high significance
    ❖ Sharpening can
    hardly be mimicked
    by other effects
    Larsen+ 2016
    Large scales Small scales

    View full-size slide

  58. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    De-lensing: TT
    ❖ Underlying CIB
    maps very simplistic:
    M545 - M857/77
    ❖ Only 40% correlated
    (ideal CIB would
    give up to 80%)
    ❖ High resolution, high
    accuracy, large scale
    CIB maps needed
    44
    Larsen+ 2016

    View full-size slide

  59. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    De-lensing: TT
    ❖ Underlying CIB
    maps very simplistic:
    M545 - M857/77
    ❖ Only 40% correlated
    (ideal CIB would
    give up to 80%)
    ❖ High resolution, high
    accuracy, large scale
    CIB maps needed
    44
    Larsen+ 2016

    View full-size slide

  60. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    CMB de-lensing of BB: The challenge
    ❖ Lensing of CMB E-modes
    leads to apparent B-modes
    ❖ One of the major
    systematics in the search
    for primordial
    gravitational waves
    45
    Courtesy A. Challinor
    Large scales Small scales

    View full-size slide

  61. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Manzotti+ (2017)
    46
    CMB de-lensing of BB
    ❖ Wiener-filter E-mode map and CIB combined in harmonic space to
    build model of lensing B-modes

    View full-size slide

  62. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    ❖ Herschel 500 micron as CIB
    template
    ❖ "No lensing" excluded at 7
    sigma
    Manzotti+ (2017)
    47
    CMB de-lensing of BB
    Large scales Small scales

    View full-size slide

  63. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    48
    Courtesy A.
    Manzotti
    CMB de-lensing of BB: Outlook
    Large scales Small scales
    Correlation with lensing

    View full-size slide

  64. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Summary on the CIB
    ❖ Template maps of Galactic dust are key to obtain large-
    scale CIB maps
    ❖ Useful by itself to constrain star formation history and
    connection of dark and luminous matter
    ❖ Best tool to-date for de-lensing CMB TT and BB, better
    CIB maps are urgently required
    49

    View full-size slide

  65. Backup slides
    50

    View full-size slide

  66. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Current CIB maps: GNILC
    ❖ Planck collaboration (2016 XLVIII), focus on removing CIB from Galactic
    dust maps
    ❖ Using the angular power spectra of the two components
    ❖ Does not agree that well on a pixel-to-pixel basis
    51

    View full-size slide

  67. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Current CIB maps: GNILC
    ❖ Planck collaboration (2016 XLVIII), focus on removing CIB from Galactic
    dust maps
    ❖ Using the angular power spectra of the two components
    ❖ Does not agree that well on a pixel-to-pixel basis
    51

    View full-size slide

  68. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Current CIB maps: GNILC
    ❖ Planck collaboration (2016 XLVIII), focus on removing CIB from Galactic
    dust maps
    ❖ Using the angular power spectra of the two components
    ❖ Does not agree that well on a pixel-to-pixel basis
    51
    0.3 0.2 0.1 0.0 0.1 0.2 0.3
    Planck XXX
    0.3
    0.2
    0.1
    0.0
    0.1
    0.2
    0.3
    GNILC
    100
    101
    102
    103

    View full-size slide

  69. Daniel Lenz, Caltech/JPL Foregrounds in observational cosmology
    Current CIB maps: GNILC
    ❖ Cross correlation with Planck CMB lensing
    ❖ Missing CIB power, especially on the largest scales
    52

    View full-size slide

  70. Residuals vs. dust
    temperature
    16 18 20 22 24 26 28
    Tdust [K]
    0.04
    0.02
    0.00
    0.02
    0.04
    (E(B V )) [mag]
    SFD Model
    16 18 20 22 24 26 28
    Tdust [K]
    MF Model
    16 18 20 22 24 26 28
    Tdust [K]
    MF SFD
    100
    101
    102
    103
    104
    105
    # data points

    View full-size slide

  71. Residuals vs. ecliptic lat
    0.050
    0.025
    0.000
    0.025
    0.050
    SFD Model [mag]
    50 0 50
    Ecliptic latitude [deg]
    0.050
    0.025
    0.000
    0.025
    0.050
    MF Model [mag]
    100
    101
    102
    103
    104
    105
    # data points

    View full-size slide

  72. 20 40 60 80 100 120
    |vcut
    LSR
    | [km s 1]
    0.0
    0.2
    0.4
    0.6
    0.8
    1.0
    Normalized RSS
    Tophat search

    View full-size slide

  73. Residuals intern
    0.004
    0.002
    0.000
    0.002
    0.004
    µ ( E(B V )) [mag]
    SFD - Model v-binned
    SFD - Model tophat
    SFD - Model simple
    0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
    NHI [1020 cm 2]
    0.000
    0.002
    0.004
    0.006
    0.008
    0.010
    ( E(B V )) [mag]
    SFD - Model v-binned
    SFD - Model tophat
    SFD - Model simple

    View full-size slide

  74. Residuals extern
    0.005
    0.000
    0.005
    0.010
    0.015
    0.020
    µ ( E(B V )) [mag]
    SFD - Model
    MF - Model
    Planck - Model
    0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
    NHI [1020 cm 2]
    0.000
    0.002
    0.004
    0.006
    0.008
    0.010
    0.012
    ( E(B V )) [mag]
    SFD - Model
    MF - Model
    Planck - Model

    View full-size slide

  75. Modeling dust foregrounds
    Model
    Residual
    Standard
    GLM
    GLM
    Standard
    Lenz+ (2016)
    58

    View full-size slide

  76. 1020 1021 1022
    NHI [cm 2]
    10 3
    10 2
    10 1
    100
    E(B V ) [mag]
    0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
    NHI [cm 2] ⇥1020
    0.00
    0.01
    0.02
    0.03
    0.04
    0.05
    0.06
    0.07
    0.08
    E(B V ) [mag]
    E(B V ) [mag] = 1.216+0.009
    0.009
    ⇥ NHI [1022 cm 2] + 0.015+0.0002
    0.0002
    [mag] = 0.02406+0.00006
    0.00006
    100
    101
    102
    103
    104
    # data points
    0
    10
    20
    30
    40
    50
    60
    70
    80
    # data points
    1020 1021 1022
    NHI [cm 2]
    10 3
    10 2
    10 1
    100
    E(B V ) [mag]
    0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
    NHI [cm 2] ⇥1020
    0.00
    0.01
    0.02
    0.03
    0.04
    0.05
    0.06
    0.07
    0.08
    E(B V ) [mag]
    E(B V ) [mag] = 1.113+0.002
    0.002
    ⇥ NHI [1022 cm 2] + 0.000+0.0001
    0.0001
    [mag] = 0.00570+0.00001
    0.00001
    100
    101
    102
    103
    104
    # data points
    0
    100
    200
    300
    400
    # data points
    The E(B-V)/NHI ratio
    Pan-STARRS E(B-V), Schlafly+ (2014) SFD E(B-V)
    Star-based Dust-based
    59

    View full-size slide