Advances and Opportunities
in Cosmology
at KASI
Eric Linder
23 September 2013
Berkeley Center for Cosmological Physics
11
Role of Observations
빨강모자 소녀
But , what big teeth you have!
Before we jump into bed with , we should
be sure it is not something more beastly.
22
Role of Theory
Cosmic acceleration and dark energy are
fundamental mysteries in understanding our
universe.
Copernican Principle / Cosmic Modesty:
• Our galaxy is not the center of the universe.
• Our particles are not the matter/energy of the
universe.
• Is our vacuum the vacuum of the universe?
• Is our gravity the gravity of the universe?
33
Are We Done?
(stat+sys)
w(z)?
z<1?
z>1?
There is a
long way to
go still to
say we have
measured
dark energy!
44
Nature of Dark Energy
Dark energy is very much not the search for
one number, “w”.
Dynamics: Theories other than  give time
variation w(z). [SN+CMB/BAO]
Degrees of freedom: Quintessence has
sound speed cs2=1. But generally w(z), cs2(z).
Is DE cold (cs2<<1), enhance perturbations?
[CMB lensing, WL]
Persistence: Is there early DE (at z>>1)?
(zCMB)~10-9 but observations allow 10-2.
[CMB, CMB x Galaxies]
Test Gravity: Expansion vs growth
[SN/BAO + CMBlens/WL/Gal]
55
1. Dynamics
Models have a diversity of
behavior, within thawing
and freezing.
But we can calibrate w by
“stretching” it: w w(a)/ a.
Calibrated parameters w0, wa.

Caldwell & Linder 2005
de Putter & Linder 2008
The two parameters w0, wa achieve 10-3 level accuracy on
This is from physics (Linder 2003).
observables d(z), H(z).
has nothing to do with a Taylor
w(a)=w0+wa(1-a) Itexpansion.
6
6
2. Persistence
We have 7 orders of magnitude of unexplored DE(zrec)=10-[2-9]
Was there early acceleration (solve coincidence)?
Was there early dark energy?
Effect of 0.1 e-fold of acceleration
Post-recombination,
peaks  left and adds ISW.
Pre-recombination,
peaks  right and adds SW.
Current acceleration unique within last
factor 100,000 of cosmic expansion!
Linder & Smith 2010
77
Early Expansion
Fit real Planck data (+WP) to any deviation from
CDM in bins of log a = [-5,-4], [-4,-3.6], [-3.6,-3.2], [3.2,-2.8], [-2.8,0].
Basically testing standard radiation/matter domination in a
model independent way.
Hojjati, Linder, Samsing 2013
Slight hint of extra
radiation energy,
i.e. Neff=3.35.
88
3. Perturbations / Microphysics
DE internal degrees of freedom can give rise to
DE perturbations (inhomogeneity).
Only significant when 1+w is not small. Since
<w(z<1)> ~ -1, this implies need early dark energy.
Perturbations only grow outside the sound horizon
~ cs/H so we need the sound speed cs<<1.
Thus persistence  early dark energy.
Degrees of freedom  cold early dark energy.
(These go together in many classes of DE theories, e.g.
Dirac-Born-Infeld or string dilaton)
99
CMB Lensing
CMB as a source pattern for weak lensing.
Probes z~1-5 effects, e.g. neutrino masses
and early dark energy.
Planck gets ~25σ for 
from CMB lensing.
10
10
CMB Lensing and Cold EDE
DE perturbations affect matter
power spectrum and so CMB
lensing. cs≠1? cvis≠0? Hu 1998
*
No EDE
EDE
*
*
Calabrese+ 1010.5612
11
11
4. Test Gravity
Test gravity in model independent way.
Gravity and growth:
Gravity and acceleration:
Are  and  the same? (yes, in GR)
Tie to observations via modified Poisson equations:
Glight tests how light responds to gravity: central to lensing
and integrated Sachs-Wolfe.
cf Bertschinger & Zukin 2008,
Song+ 2011
Gmatter tests how matter responds to gravity: central to
growth and velocities ( is closely related).
12
12
Extending Gravity
Interesting recent theories extending gravity for cosmic
acceleration often have shift symmetries (ϕ  ϕ+c) and
higher order kinetic terms (ϕμϕμ, related to higher dimensions or
massive gravity).
From Horndeski general scalar-tensor theory,
Charmousis+ 2011 found “Fab 4” unique self tuning terms.
De Felice, Linder 2012 promote to nonlinear, mixed function.
Appleby,
ρ
Fab 5 accelerates, has
tracker, dS attractor, no
extra dof! – and self
tuning.
It makes  invisible!
also see Padilla & Sivanesan 2013
H2
13
13
Chasing Down Cosmic Acceleration
How can we measure dark energy in detail?
A strong program of techniques is in place, but
new, complementary probes offer exciting
prospects.
• Measuring distance to 1%
with strong lensing time delays
• Measuring growth and gravity
to %-level with redshift space
distortions
14
14
Strong Lensing Time Delays
Strong gravitational lensing creates multiple images
(light paths) of a source.
When the source is variable (quasar / AGN), we can
measure the time delays between the images. This
probes the geometric path difference (cosmology)
and the lensing potential (dark matter).
Key parameter is a distance ratio, the time delay
distance
15
15
Time Delays + Supernovae
Lensing time delays give superb complementarity
with SN distances plus CMB.
(Planck)
T to 1% for
zl=0.1, 0.2,… 0.6
SN to 0.02(1+z)mag
for z=0.05, 0.15... 0.95
Factor 4.8 in area
Ωm to 0.0044
h to 0.7%
w0 to 0.077
wa to 0.26
immunizes vs curvature
Linder 2011
16
16
Strong Lensing Distance Surveys
Best current distances at 5% accuracy (16 systems known, 2-5
at 5%). 5 year aim: 25 systems, 5% accuracy = 150 orbits HST.
Long term: 1% distances.
Need 1) high resolution imaging for lens mapping / modeling,
2) high cadence imaging, 3) spectroscopy for redshift, lens
velocity dispersion, 4) wide field of view for survey.
Synergy: HST/GMT+ LSST/Euclid. Only low redshift
z<0.6 needed for lenses.
KASI Opportunity –
EACOA partner ASIAA has
lens modeling expertise
- Time domain monitoring (1-2m telescope, CosmoGrail)
- AGN flux and time variation expertise
- GMT (AO): highest resolution ground telescope
- Cosmology expertise on joint probe analysis
17
17
Measuring Time Delays
One of the challenges is measuring time delays
between images in the presence of 1) noise, 2) gaps,
3) variability, 4) microlensing.
We use Gaussian Process statistics to find a family
of light curves fitting the data, with correlations. See
Holsclaw+ 2010a,b,2011; Kim, Linder, Shafieloo 2012,2013 for
cosmology.
applications to
18
18
LSST Data Challenge
Hojjati, Kim, Linder 2013
Blind fits:
Truth:
Real data: accurate and more precise than literature.
Precision: ~2x improvement (1-2%)
Accuracy: within 0.35σ (blind challenge)
LSST Data Challenge: “Ladder of Evil”
LSST will be a lens finding factory, ~8000 lensed AGN.
Strong lensing probes dark energy and dark matter.1919
Higher Dimensional Data
Cosmological Revolution:
From 2D to 3D – CMB anisotropies to
tomographic surveys of density/velocity field.
Data, Data, Data – CMB maps l2~10M modes;
BOSS maps k3V~0.4M modes; DESI 15M modes
20
Redshift Space Distortions
Redshift space distortions (RSD) map velocity field
along line of sight. Gets at growth rate f, one less
integral than growth factor (like H vs d).
Ωm =
Linder 2005
gravitational
growth index 
Hume Feldman
21
21
Beyond Linear
Kaiser formula inaccurate
Even monopole (average
over RSD) is poor.
Anisotropic redshift
distortion hopeless –
without better theory.
k (h/Mpc)
Major approaches are high order perturbation theory,
computational simulations – or a new combination of
the two.
22
22
High Order Perturbation Theory
Putting a rigorous formalism into place:
23
23
High Order Perturbation Theory
Applying perturbation theory to simulation and data:
Density growth
Velocity growth
BAO
RSD
Song, Okumura, Taruya 2013
Distance
Hubble parameter
24
24
Simulations + Analytics
Simulations allow calibration of RSD into nonlinear
regime, where most of the information is.
Perturbation theory informs analytic form for
linear  nonlinear transformation.
Ptrue(k,)=F(k) Plin(k,)
Kwan, Lewis, Linder 2012
Use simulations to
calibrate A, B, C(k,z).
k
25
25
Simulations to Fit RSD
First results: KLL analytic reconstruction form is
accurate to 2-5% out to k=0.5 h/Mpc for z=0-2.
(Ppred-Psim)/Psim
Vallinotto & Linder 2013
26
26
Dark Energy Spectroscopic Instrument
DESI on the Mayall Telescope, Kitt Peak
3D maps of z=0-3.5
20M galaxies, quasars
KASI Opportunity –
- Spectrocopic survey in optical/NIR
- Fiber optic engineering and testing
- Cosmology expertise on BAO, RSD (growth/gravity)
- High performance computing (KISTI partner?)
27
27
Dark Energy Spectroscopic Instrument
1M LyA QSO
2.5M QSO
18M ELG
4M LRG
100% dark time
DOE granted “mission
need” 2012
$2.1M Moore grant 2013
CDR milestone Jan 2014
Survey start 2018
KASI Opportunity –
- Spectrocopic survey in optical/NIR
- Fiber optic engineering and testing
- Cosmology expertise on BAO, RSD (growth/gravity)
- High performance computing (KISTI partner?)
28
28
Testing Gravity
Model independent tests of gravity: two functions, at
high/low z, high/low k (8 tests). Simultaneous fit for gravity,
expansion (w0, wa), galaxy bias (27 bins).
DESI+Planc
k
Daniel & Linder 2013
Fit all
Fix to 
Fix to b~1/D
29
29
Summary
Dark energy is not the search for one number “w”.
Explore dynamics, degrees of freedom, persistence.
Measuring the growth history, testing growth vs
expansion, probes dark energy, gravity, dark matter.
Strong program in place + new probes
• Strong lensing time delays [HST/GMT; LSST]
• Redshift space distortions [DESI]
Cosmology at KASI well positioned to use its
existing strengths and develop new ones.
Science overlap with several groups: GMT, AGN, IR,
spectroscopy, fibers. Partnering: ASIAA, KISTI.
30
30