Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology Before we start… ❖ Talk to me about: ❖ Galactic HI ❖ Observational multiphase ISM ❖ CIB/CMB component separation, de-lensing ❖ Dust/Reddening ❖ HI intensity mapping (simulations) ❖ Machine learning, Bayesian models, python, Mac OS 2
Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology Before we start… ❖ Talk to me about: ❖ Galactic HI ❖ Observational multiphase ISM ❖ CIB/CMB component separation, de-lensing ❖ Dust/Reddening ❖ HI intensity mapping (simulations) ❖ Machine learning, Bayesian models, python, Mac OS 2
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' Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 3
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 ❖ Dust and synchrotron foregrounds in CMB data Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 3
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 Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 5
Mapping E(B-V) Direct approach ❖ Find many sources with known spectrum (e.g. stars, passive galaxies) ❖ Measure spectra, attribute differences to dust ❖ E.g. Schlafly+ (2014), Green+ (2015) using Pan-STARRS ❖ Limited by modeling accuracy and sensitivity Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 6
Mapping E(B-V) Direct approach ❖ Find many sources with known spectrum (e.g. stars, passive galaxies) ❖ Measure spectra, attribute differences to dust ❖ E.g. Schlafly+ (2014), Green+ (2015) using Pan-STARRS ❖ Limited by modeling accuracy and sensitivity Indirect approach ❖ Measure dust optical depth, linearly related to E(B-V) ❖ Schlegel, Finkbeiner, Davis (SFD, 1998) still state-of-the art ❖ FIR emission may have contributions from Zodiacal Light and unresolved galaxies, also needs dust temperature correction Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 6
Mapping E(B-V) Direct approach ❖ Find many sources with known spectrum (e.g. stars, passive galaxies) ❖ Measure spectra, attribute differences to dust ❖ E.g. Schlafly+ (2014), Green+ (2015) using Pan-STARRS ❖ Limited by modeling accuracy and sensitivity Indirect approach ❖ Measure dust optical depth, linearly related to E(B-V) ❖ Schlegel, Finkbeiner, Davis (SFD, 1998) still state-of-the art ❖ FIR emission may have contributions from Zodiacal Light and unresolved galaxies, also needs dust temperature correction Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 6
HI emission as basis for E(B-V) ❖ Gas and dust are well-coupled in the ISM ❖ Perform an SFD-like analysis to convert HI emission to E(B-V) ❖ Resulting maps free from errors due to dust temperature, Zodi, and extragalactic emission ❖ Limited by non-HI gas along the line of sight Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 7
Dust systematics -0.01 0.01 SFD - PG10 [mag] -0.01 0.01 SFD - Model [mag] Based on extragalactic sources Based on galactic HI Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 11
The Cosmic Infrared Background Made up from dust in galaxies at z=1-3 Lagache+ (2002) Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 13
The Cosmic Infrared Background Made up from dust in galaxies at z=1-3 Lagache+ (2002) Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 13
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 Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 14
❖ 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 Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 15
❖ 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 Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 15
How to obtain CIB maps? ❖ Galactic thermal dust and CIB dust dominate on large scales at ~200 to 1000 GHz ❖ How to disentangle them? Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 16
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? Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 16
Correlation of dust and gas HI Dust ❖ Linear relation to first order, but better model required to get to CIB levels Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 17
Modeling dust foregrounds • Generalised linear model (GLM) • Radial Velocity I = X i ✏iTi B Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 19
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 | Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 20
CIB: Access large scales ❖ Sliding window is moved across the sky ❖ Model is evaluated for each position, yields map of parameters and CIB values Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 21
CIB: Access large scales ❖ Sliding window is moved across the sky ❖ Model is evaluated for each position, yields map of parameters and CIB values Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 21
Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology The challenge ❖ Lensing of CMB E- modes leads to apparent B-modes ❖ De-lensing of this effect through internal algorithms or tracers of the large-scale structure 24 Courtesy A. Challinor
The CIB as cosmological probe … of large scale structure to de-lens CMB maps Manzotti+ (2017) Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 25
❖ Herschel 500 micron as CIB template ❖ "No lensing" excluded at 7 sigma … of large scale structure to de-lens CMB maps Manzotti+ (2017) The CIB as cosmological probe Daniel Lenz, Caltech/JPL Foregrounds in observational Cosmology 26
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 31
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 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 32
daniellenz
ishiijunpei
genomethica
koro485
signer
norihirosuzuki
konakalab
mitsuharu
ecoopnet
miyayou
dasaptaerwin
leeswijzer
sferik
mraible
aki_iinuma
tammielis
frogandcode
sachag
eileencodes
shlominoach
jlugia
samanthasiow
matthewcrist
andyhume