Transcript CMB Cosmological Parameter Extraction

Early & Late Inflation Theory
cf. CMB+LSS @ CITA
Concordance (pre-WMAP2++) & discordant notes. The “basic
parameters”: LCDM with power law index and no/little GW is concordant. Some
discord .. always. Extension to larger parameter lists.
Are current discordant notes statistically significant, systematics,
astrophysics, fundamental physics? Add parameters: theory priors – cost
function, baroqueness, taste? r_ts (<.72/<.36??), k_BSI, dn_s/dlnk, size_U/chi_U,
subdominant (isoc/adiab), z_reh (+VMOs?) {e.g. Vaudrevange, Lewis, McDonald @
CITA/PI focus gp; soon Nolta WMAP2}
Resolving (or celebrating) discord with planned experiments. break
degeneracies of cosmic parameters planned CMB+ext, with ext= z-surveys, cl-gp
surveys (Sunyaev-Zeldovich/Opt/X/weaklens), weak lensing LSS surveys, SN
CMB High-L frontier TT: near term, cbipol, boom2Kpol, acbar
forecasts. Long term Apex, ACT, SPT, Quiet forecasts.
CMB Polarization EE/BB, High & Low-L: CBI, BOOM2K , DASI,
CAPMap, WMAP2, BICEP, QUaD, Polarbear, QUIET, AMiBA, Planck, … CMBpol
Anomalies: Systematics, Statistics,
AstroPhysics or Fundamental Physics?
“anomalies” @ low L 2,3; ~20-30; check@200, > 600?
ET checkmarks 2, 22, 222, 2222
Issues: L=2,3 how low is the probability? Glitches? non-WMAP data
e.g. Acbar/CBI calibration
CBI anomalous power @ L > 2000, Sunyaev Zeldovich effect in the
cosmic web is plausible if s8 if ~ 0.9 (nonlinear gasdynamics)
Statistical isotropy broken on large scales?
Jan04
CMB data
cf.
good-fit
LCDM
uniformacceleration
model
String Theory Landscape @ Stanford
Perhaps 10100
different vacua
single field inflation vs. WMAP1+ext+ext
WMAP1 Peiris et al. 2003
Each point is a “viable slow-roll” model, able to sustain inflation for sufficient e-foldings to
solve the horizon problem and make the Universe (nearly) flat BUT highly restricted.
Monte Carlo simulations using “flow-equations” (Kinney 2002). slow-roll priors
are artificial, highly restrictive & misleading about allowed inflationary phase
space. Equivalent to a power series expansion of order 5, 6, 8 of histories.
CITA approach: efficiently sampled inflation acceleration histories (trajectories).
Physics priors: NONE - all are allowed (constrain by taste, e.g. allowed bumpiness).
Only data can decide, includes very weak “priors” like “anthropic data” as well as
high precision CMB+LSS. Monte Carlo simulation of acceleration histories
String Theory Landscape & Inflation++ Phenomenology for CMB+LSS
running index as simplest breaking (back to early 80s), radically broken
scale invariance, 2+-field inflation, isocurvatures, Cosmic
strings/defects, compactification & topology, & other baroque add-ons.
subdominant
String/Mtheory-motivated, extra dimensions, brane-ology, reflowering of
inflaton/isocon models (includes curvaton), modified kinetic energies, kessence, Dirac-Born-Infeld [sqrt(1-momentum**2), “DBI in the Sky”
Silverstein etal 2004], etc.
any
14 std inflation
parameters acceleration
trajectory will
do??
+ many many
more e.g.
q (ln Ha)
“blind” search
H(phi,…)
for patterns in
the primordial
V(phi,…)
power
Measure??
spectrum
anti-baroque
prior
KKLT, KKLMMT
Potential of the Hybrid D3/D7
Inflation Model
Running of the Spectral Index : Parameter Fits
[Spergel et al. ApJS, 148, 175]
[Tegmark et al. astro-ph/0310723]
[Bond, Contaldi & Pogosyan astro-ph/0310735]
[CBI VII, Readhead et al. astro-ph/0402359]
CMB data
only
[dns/dlnk]
<0 in
Jan03,
Mar03 &
Mar04
CMB
+ weak
lens s8
PLANCK vs. WMAP4yr + Ground based telescopes
(circa ~ 2008)
• WMAP 4yr
• + Ground-based telescopes pre-Planck ACT/SPT-like (bolometers) ; QUIET (HEMT
arrays). Coverage assumed; ~few % of the sky (1000 sq deg); polarization included
•PLANCK (2007+)
END
No running: dn/dlnk~0
Damped Lyman alpha
system effect at small
scales + increased
WMAP errors at large
scales of Slosar &
Seljak 04
Systematic error budget? ..
McDonald etal, Haehnelt etal
acceleration
trajectories,
“featured”
potentials,
expansions about
uniform q(T) (aka
slow-roll-ology in
the HamiltonJacobi/stochastic
inflation
formulation –
works OK), … &
the nonlocal
WKB++
approximation
(Habib etal
02,04)
B+Contaldi+Kofman+Souradeep+Vaudrevange+