The Epoch of Reionization: Observational & Theoretical Topics

Lecture 1 Lecture 2 Lecture 3 Lecture 4

Current constraints

on Reionization Physics of the

21cm probe EoR radio

experiments Expected

Scientific outcome

i. CMB Polarization.

ii. Lyman-a forest data.

iii. Opacity of ionizing photons (`Bolton et al).

iv. Temperature

evolution (Theuns etal, Haiman&Hui) v. Soft Xray BG

(Dijkstra ..) vi.IR BG (HESS

results) vii.HST WPC3

results

i. Basic Formulae (Field 1958) ii. Excitation

mechanisms (Ly- a, collisions,..) iii. Global evolution

of the spin temp.

iv. Patchy evolution v. Simulation results

i. Current & future experiments.

ii. Key parameters in experiments.

iii. Observational issues: uv coverage, foregrounds, ionosphere,

instrument, noise.

iv. Extraction issues.

v. Calibration.

vi.Polarization.

i. Cosmology:

Density field, ionization frac.

Redshift distort, power spect.

ii. First sources iii. Ionization history iv. Dark ages and

history of spin temperature.

v. The future.

Lecture 2:

Physics of the 21cm line probe

i. Historic overview

ii. Basic Formulae (Field 1958) iii.Excitation mechanisms (Ly-a,

collisions,..)

iv. Global evolution of the spin temp.

v. Patchy evolution

vi. Simulation results



CMB (integral constraint)



Redshifted 21 cm emission (absorption)



21 cm forest at high z



Gamma ray bursts: How many we should have to constrain reionization?



Luminosity function of first objects, e.g., Galaxies:

Recent results from the new WFC3 aboard HST.

Key Probes of Reionization



Background detections: IR, soft x-ray.



Lyman-a absorption system:

ionization, metallicity, thermal history, UV background,

proximity effect.



Lyman alpha emitters



Metals at high redshift.



Using the local volume to

study reionization.

Historic overview

 H.C. van de Hulst (inspired by J. Oort) showed the potential of the 21 cm transition in astronomy - 1945

 The frst astronomical observation of the 21 cm: H.I. Ewen & E.M.

Purcell (1951, Nat. 168, 356)

C.A. Muller & J.H. Oort (1951, Nat. 168, 357-8)

 Excitation mechanism Wouthuysen (1952). Field (1958, 1959) gave the proper framework.

 Importance for cosmology was inspired by Zel'dovich's top down scenario.

 Scott & Rees (1992) pointed out that a signal could detected from high z 21 cm.

 Madau, Meiksin & Rees (1997) were the frst to consider the interplay between the frst sources and the 21 cm transition.

 Over the years many observational attempts failed. It is only now that we think that we can observe high redshift 21 cm radiation.

5

21-cm Physics

1420 MHz Mechanisms

??

Lifetime of ~10 Myrs

The 21 cm transition

• The value of the T

is given by:

n₀, g₀ n₁, g₁

21 cm

Field 1958 Madau et al 98

Ciardi&Madau 2003

Lyman-a Coupling

• The Wouthuysen- Field effect, also known as Lyman- alpha pumping.

Dominant in both in the case of stars and Black- holes, due to photo and collisional excitations,

respectively.

Wouthuysen 1952 Field 1958

Collisional Coupling



H-H collisions that excite the 21 cm

transition. This interaction proceeds through electron exchange.



H-e collisions. Especially important around primordial X-ray sources (mini-quasars).

–

This effect might also excite Lyman-alpha transition which adds to the T

- T

decoupling effciency.

dT

, The Brightness Temperature

T

T

T

Where the optical depth is given by:

A

= 2.85x10

s

is the spontaneous emission coefficient.

N

is the column density of HI; 4 accounts for fraction in singlet state f(n) is the line profile.

An accurate calculation of the optical depth at a given redshift, which

takes into account line profile broadening due to Hubble expansion and

casts the relation in terms of number density, yields:

δT _b : Brightness temperature

Cosmology Astrophysics



The Interpretation might be very complicated



Notice that the signal in absorption can be

much smaller

The Global evolution of the Spin Temperature

Loeb & Zaldarriaga 2004,Pritchard & Loeb 2008, Baek et al. 2010, Thomas & Zaroubi 2010

At z~20 T

is tightly coupled to T

. In order to observe the 21 cm radiation

decoupling must occur.

Heating much above the CMB temp. and

decoupling do not

necessarily occur together.

This drives the Compton heating rate to almost zero Compton heating

rate

Compton cooling time

The redshift of thermal decoupling is about 200

(proper calculation could be done with the publicly available code RECFAST)

The Spin Temperature Prior to the EoR

Loeb &

Zaldarriaga 04

Only feasible from the Moon

Ionization sources

Mean free path

Bound-free Cross section

n

= 2.2 x 10

cm

(1+z)



^{= 6 x 10}

^cm

E

= 13.6 eV

At z = 9: For E = E

l

 ≈ 2 kpc comoving

For E = 1 keV l

 ≈ 1 Mpc comoving

X-ray photons UV photons

Large cross section but ejected electron has low energy

Low cross section but ejected electron has high energy

e^-

e^-

The fraction of photon energy that goes to reionization, heating and excitation is roughly 1:1:1 as calculated with Monte-Carlo radiative transfer code by Shull & van

Steenberg (1986) and Valdes et al. 2009.

The signal: Stars vs. Miniqsos

Thomas &

Zaroubi 2008

Kinetic temperature is greatly

heated just beyond the HII region, but further out it has been

adiabatically cooled.

21cm absorption strongly dominates over the inner emission core

Redshift x-ray source

Thomas & Zaroubi 2008

Simulations of the EoR



Cosmological Hydro simulations:

1- High enough resolution to resolve halos in which ionization sources form. 2- Span Large Scales as well as small scales, especially since designed arrays have small 1' res. 3- In certain cases DM only simulations are suffcient.

 Out of equilibrium Radiative Transfer:

1- Source and their fux. 2- Ionization of H and He (not always done). 3- Heating due to the radiative processes. 4- Spin temp decoupling (Lya RT).

 It is very diffcult to account for all the physical aspects of the problem and approximations are normally made.

Results from 3D RT

Iliev et al. 2008

At half ionization the signal rms is about 8mK

~12mK

Results from approximate methods

Thomas & zaroubi 2008

Full vs. approximate simulations

 Full 3D RT simulations are more accurate but

computationally expensive.

They provide crucial insight about the physical processes (especially on small scales).

 Approximate methods are less accurate but easier to

produce and allow for an

exploration of the parameters space. This is especially

important for interpretation of the data

Thomas et al 2009

Spin Temperature issues

In case the spin temp. is of the order the CMB temp.

or smaller an absorption signature is expected at high redshifts.

Thomas & Zaroubi 2010 See also Baek et al. 2010

The 21 cm forest

Simulated spectrum from 100 MHz to 200 MHz of a source with S₁₂₀

= 20 mJy at z=10 using the Cygnus A spectral model and SKA noise

Summary



21 cm line is a very promising probe of the EoR and the Dark Ages.



It tracks the evolution of reionization and the thermal history of the IGM in time and space.



The signal is of the order of 10 mK in

emission and 100 mK in absorption.