Shedding light on the epoch of reionization with the 21cm signal - - PowerPoint PPT Presentation

Anne Hutter

Catherine Watkinson, Jacob Seiler, Cathryn Trott, Pratika Dayal, Darren Croton, Manodeep Sinha,

Shedding light on the epoch of reionization with the 21cm signal

ARC CENTRE OF EXCELLENCE FOR ALL SKY ASTROPHYSICS IN 3D

Anne Hutter Epoch of Reionization 2

The Epoch of Reionization

What is the reionization history, the evolution of ⟨χHI⟩? What is the ionization topology, the evolution of the sizes and location of the ionized regions? What are the sources of reionization and what is their escape fraction of ionizing photons into the IGM? How does reionization afgect galaxy formation & evolution?

Anne Hutter Epoch of Reionization 3

The Epoch of Reionization

What is the reionization history, the evolution of ⟨χHI⟩? What is the ionization topology, the evolution of the sizes and location of the ionized regions? What are the sources of reionization and what is their escape fraction of ionizing photons into the IGM? How does reionization afgect galaxy formation & evolution?

Anne Hutter Epoch of Reionization 4

Observational constraints on reionization

➢ Cosmic Microwave Background

• ptical depth τ = 0.054 ± 0.007

➢ QSO absorption spectra ➢ Lyman-α emitter luminosity

function & clustering

➢ GRB afterglow spectra

Planck 2018

Davies+ 2018 Konno+ 2018 IGM damping wing in QSO absorption spectra during reionization

Anne Hutter Epoch of Reionization 5

Lyman α emitters (LAEs) in the intergalactic medium

fesc = escape fraction of ionizing photons ( λ<912Å )

ISM HI

Lyman α photons emitted (recombination radiation) ISM = interstellar medium IGM = intergalactic medium

IGM galaxy

Hutter+ 2014

Anne Hutter Epoch of Reionization 6 fesc = escape fraction of ionizing photons ( λ<912Å )

galaxy ISM HI

Lyman α photons emitted (recombination radiation) ISM = interstellar medium IGM = intergalactic medium

IGM

neutral

ionized

Hutter+ 2014

Lyman α emitters (LAEs) in the intergalactic medium

Anne Hutter Epoch of Reionization 7 fesc = escape fraction of ionizing photons ( λ<912Å ) fα = escape fraction of Lyman α photons ( λ=1216Å )

galaxy ISM HI

Lyman α photons emitted (recombination radiation)

dust

scattering absorption

ISM = interstellar medium IGM = intergalactic medium

IGM

neutral

ionized

Hutter+ 2014

Lyman α emitters (LAEs) in the intergalactic medium

Anne Hutter Epoch of Reionization 8 fesc = escape fraction of ionizing photons ( λ<912Å ) fα = escape fraction of Lyman α photons ( λ=1216Å )

galaxy ISM HI

Lyman α photons emitted (recombination radiation)

dust

scattering absorption

ISM = interstellar medium IGM = intergalactic medium

IGM

neutral redshifted out of resonance & transmitted through IGM

ionized

absorption by HI

Hutter+ 2014

Lyman α emitters (LAEs) in the intergalactic medium

LAE selection criteria: Lα

• bs >1042 erg s-1

EW > 20 Å

Anne Hutter Epoch of Reionization 9 fesc = escape fraction of ionizing photons ( λ<912Å ) fα = escape fraction of Lyman α photons ( λ=1216Å )

galaxy ISM HI

Lyman α photons emitted (recombination radiation)

dust

scattering absorption

ISM = interstellar medium IGM = intergalactic medium

IGM

neutral redshifted out of resonance & transmitted through IGM

ionized

absorption by HI

Hutter+ 2014

Lyman α emitters (LAEs) in the intergalactic medium

LAE selection criteria: Lα

• bs >1042 erg s-1

EW > 20 Å

intrinsic luminosity dust attenuation IGM attenuation

Anne Hutter Epoch of Reionization 10

Lyman α emitters as probe of reionization

Konno+ 2018 dust increases

Anne Hutter Epoch of Reionization 11

Lyman α emitters as probe of reionization

Konno+ 2018 Ionizing escape fraction increases dust increases

Anne Hutter Epoch of Reionization 12

Lyman α emitters as probe of reionization

Konno+ 2018 HI in IGM increases Ionizing escape fraction increases dust increases

Degeneracy between the ionizing escape fraction, dust attenuation & reionization (Hutter+ 2014)

Anne Hutter Epoch of Reionization 13

Lyman α emitters as probe of reionization

Konno+ 2018 HI in IGM increases Ionizing escape fraction increases dust increases

Degeneracy between the ionizing escape fraction, dust attenuation & reionization (Hutter+ 2014)

z~8.7

z~8.7 EGSY8p7 (Zitrini+2015) z~7.7 EGS-zs8-1 (Oesch+ 2015) z~7.5 EGS-zs82 (Roberts-Borsani+ 2016)

z~7.7 z~7.5

Anne Hutter Epoch of Reionization 14

21cm SIGNAL

(maps)

Ionization topology with the 21cm signal

n1 n0 Distribution of atoms in states F=0 and F=1 is given by the spin temperature Ts & controls intensity of the 21cm signal During the Epoch of Reionization the 21cm signal traces the neutral hydrogen density. Ionized regions appear as holes.

Anne Hutter Epoch of Reionization 15 cross correlation function ξ21cm,LAE distance [h-1 Mpc]

⟨χHI⟩ decreases ⟨XHI⟩ = 0.5 ⟨XHI⟩ = 0.25 ⟨XHI⟩ = 0.1

photoionization rate increases

Hutter+ 2017

Constraining reionization with 21cm - LAE cross correlations

see also Vrbanec+2016, 2019, Sobacchi+ 2016, Heneka+2017, Kubota+2018

21cm-LAE cross correlations trace:

➢ IGM ionization state (amplitude) ➢ size of ionized regions around

LAEs (turn over) 21 cm galaxies

Lidz+ 2009

Hydro + RT 80h-1 Mpc box

Anne Hutter Epoch of Reionization 16

1 10 1 10 1 10 r [h-1 Mpc] r [h-1 Mpc]

⟨XHI⟩ = 0.5 ⟨XHI⟩ = 0.25 ⟨XHI⟩ = 0.1

δ P21 ,LAE

2

(k) = 2 P21, LAE

2

(k) + 2 [P21(k) + σ21

2 (k)] [P LAE(k) + σLAE 2

(k)]

21cm sample variance LAE sample variance thermal noise shot noise

Survey luminosity limit Survey volume

decreases decreases

r [h-1 Mpc] 1 10 Hutter+ 2018

Best survey parameters for detecting 21cm-LAE cross correlations with SKA

Anne Hutter Epoch of Reionization 17

1 10 1 10 1 10 r [h-1 Mpc] 1 10 r [h-1 Mpc] r [h-1 Mpc]

δ P21 ,LAE

2

(k) = 2 P21, LAE

2

(k) + 2 [P21(k) + σ21

2 (k)] [P LAE(k) + σLAE 2

(k)]

21cm sample variance LAE sample variance thermal noise shot noise

Survey luminosity limit Survey volume

decreases decreases SILVERRUSH HSC ⟨XHI⟩ = 0.5 ⟨XHI⟩ = 0.25 ⟨XHI⟩ = 0.1 z = 6.6 Δz = 0.1 HSC: 1.8 deg2 SILVERRUSH: 21 deg2

Hutter+ 2018

Best survey parameters for detecting 21cm-LAE cross correlations with SKA

Anne Hutter Epoch of Reionization 18

1 10 1 10 1 10 r [h-1 Mpc] 1 10 r [h-1 Mpc] r [h-1 Mpc]

δ P21 ,LAE

2

(k) = 2 P21, LAE

2

(k) + 2 [P21(k) + σ21

2 (k)] [P LAE(k) + σLAE 2

(k)]

21cm sample variance LAE sample variance thermal noise shot noise

Survey luminosity limit Survey volume

faint bright intermediate Lα = 1041-42 erg s-1 Lα = 1042-43 erg s-1 Lα > 1043 erg s-1 HSC SILVERRUSH

Hutter+ 2018

Best survey parameters for detecting 21cm-LAE cross correlations with SKA

Anne Hutter Epoch of Reionization 19

FoV [deg2] 1 10 V [Mpc3] V [Mpc3] V [Mpc3] FoV [deg2] FoV [deg2] 1 10 1 10

With current instruments shallow large FoV surveys are achievable FoV > 20 deg2 and Lα > 8x1042 erg s-1

Hutter+ 2018, 2019

Best survey parameters for detecting 21cm-LAE cross correlations with SKA

Anne Hutter Epoch of Reionization 20

The Epoch of Reionization

What is the reionization history, the evolution of ⟨χHI⟩? What is the ionization topology, the evolution of the sizes and location of the ionized regions? What are the sources of reionization and what is their escape fraction of ionizing photons into the IGM? How does reionization afgect galaxy formation & evolution?

Anne Hutter Epoch of Reionization 21

The Epoch of Reionization

What is the reionization history, the evolution of ⟨χHI⟩? What is the ionization topology, the evolution of the sizes and location of the ionized regions? What are the sources of reionization and what is their escape fraction of ionizing photons into the IGM? How does reionization afgect galaxy formation & evolution? gas

star formation

GALAXY

Escape fraction

• f ionizing photons

GALAXY

Shedding light on reionization with 21cm 22

Bouwens+ 2019

The sources of reionization

Star-forming galaxies can provide enough HI ionizing photons to reionize the

• Universe. e.g. Finkelstein+ 2019,

Robertson+2015, Mutch+ 2016, Seiler+ 2019

Contribution from AGN is probably small.

e.g. Weigel+ 2015, Parsa+2017, Onoue+2017, Kulkarni+ 2019, Qin+ 2017, Yoshiki+ 2018, Trebitsch+ 2018, Mitra+ 2018

UV luminosity function

Shedding light on reionization with 21cm 23

Bouwens+ 2019

The sources of reionization

UV luminosity function

˙ Nion = f esc ξion ρUV

The number of ionizing photons is given by

Escape fraction

• f LyC photons

LyC photon production effjciency UV luminosity density of galaxies

?

Star-forming galaxies can provide enough HI ionizing photons to reionize the

• Universe. e.g. Finkelstein+ 2019,

Robertson+2015, Mutch+ 2016, Seiler+ 2019

Contribution from AGN is probably small.

e.g. Weigel+ 2015, Parsa+2017, Onoue+2017, Kulkarni+ 2019, Qin+ 2017, Yoshiki+ 2018, Trebitsch+ 2018, Mitra+ 2018

Shedding light on reionization with 21cm 24 The number of ionizing photons is given by

The escape fraction of ionizing photons into the IGM

Bouwens+ 2016, Harikane+ 2018, Lam+ 2019, Naidu+ 2019

˙ Nion = f esc ξion ρUV

Escape fraction

• f LyC photons

LyC photon production effjciency UV luminosity density

• f galaxies

Naidu+ 2019

ξion: inferred from Hα emission line measurements at z~4-5 ρ

UV: inferred from UV luminosity

functions at z=4-10

Anne Hutter Epoch of Reionization 25

The ionizing escape fraction & the ionization topology

BUT: realistically the escape fraction of ionizing photons will depend on the physical processes and gas distributions in the galaxies

gas

star formation

GALAXY

Escape fraction

• f ionizing photons

GALAXY

Paardekooper+ 2015, Kimm+ 2017, 2019, Trebitsch+ 2018, Seiler+ 2018

Anne Hutter Epoch of Reionization 26

The ionizing escape fraction & the ionization topology

BUT: realistically the escape fraction of ionizing photons will depend on the physical processes and gas distributions in the galaxies

low Mhalo high Mhalo

Anne Hutter Epoch of Reionization 27

The ionizing escape fraction & the ionization topology

BUT: realistically the escape fraction of ionizing photons will depend on the physical processes and gas distributions in the galaxies

low Mhalo high Mhalo

fesc = constant

Anne Hutter Epoch of Reionization 28

The ionizing escape fraction & the ionization topology

BUT: realistically the escape fraction of ionizing photons will depend on the physical processes and gas distributions in the galaxies

low Mhalo high Mhalo

fesc increases with Mhalo

Anne Hutter Epoch of Reionization 29

The ionizing escape fraction & the ionization topology

BUT: realistically the escape fraction of ionizing photons will depend on the physical processes and gas distributions in the galaxies

low Mhalo high Mhalo

fesc decreases Mhalo

Anne Hutter Epoch of Reionization 30

The Epoch of Reionization

What is the reionization history, the evolution of ⟨χHI⟩? What is the ionization topology, the evolution of the sizes and location of the ionized regions? What are the sources of reionization and what is their escape fraction of ionizing photons into the IGM? How does reionization afgect galaxy formation & evolution?

Shedding light on reionization with 21cm

The ionization topology

fesc ∝ fej fesc ∝ SFR fesc = const.

Ionized bubbles have similar sizes

➢ fesc increases with SFR

(↑ with Mh)

➢ fesc is constant. ➢ fesc increases with the

ejected gas fractions (↓ with Mh) gas

star formation

GALAXY

Escape fraction

• f ionizing photons

GALAXY

Seiler, Hutter+ 2019

160 cMpc RSAGE: semi-numerical model

• f galaxy evolution & reionization

Shedding light on reionization with 21cm

The ionization topology

fesc ∝ fej fesc ∝ SFR fesc = const.

Ionized bubbles have similar sizes

Hutter+ 2019, arXiv 1907.04342

f

e s c

= c

• n

s t f

e s c

∝ S F R

Ionized regions become larger from fesc ∝ fej to fesc ∝ SFR 160 cMpc

Shedding light on reionization with 21cm

The ionization topology

fesc ∝ fej fesc ∝ SFR fesc = const.

Ionized bubbles have similar sizes

Hutter+ 2019, arXiv 1907.04342

f

e s c

= c

• n

s t f

e s c

∝ S F R

Neutral regions become also larger from fesc ∝ fej to fesc ∝ SFR 160 cMpc

Shedding light on reionization with 21cm 34

More information with higher-order statistics?

The 21cm power spectrum only traces the Gaussian part. Analyzing non-Gaussianities with the 21cm BISPECTRUM during reionization The 21cm signal from reionization is non-Gaussian.

Shedding light on reionization with 21cm 35

21cm bispectra tracing ionization topology?

(2π)

3 B( ⃗

k1, ⃗ k2, ⃗ k3) δD( ⃗ k1+ ⃗ k2+ ⃗ k3)=⟨Δ( ⃗ k1) Δ( ⃗ k2) Δ( ⃗ k3) ⟩

Bispectrum is the Fourier transform of the 3-point correlation function:

contributes only when ⃗ k1+ ⃗ k2+ ⃗ k3=0

|k3| increases

k

1

k

1

k

2

k

2

k

3

k

3

k

1

k

2

k

3

θ

e q u i l a t e r a l s t r e t c h e d s q u e e z e d

• 1

≤ c

• s

θ <

• .

5

• .

5 < c

• s

θ ≤ 1

θ θ

c

• s

θ =

• .

5

Watkinson+2017

Shedding light on reionization with 21cm 36

21cm bispectra tracing ionization topology?

(2π)

3 B( ⃗

k1, ⃗ k2, ⃗ k3) δD( ⃗ k1+ ⃗ k2+ ⃗ k3)=⟨Δ( ⃗ k1) Δ( ⃗ k2) Δ( ⃗ k3) ⟩

Bispectrum is the Fourier transform of the 3-point correlation function:

contributes only when ⃗ k1+ ⃗ k2+ ⃗ k3=0

|k3| increases

k

1

k

1

k

2

k

2

k

3

k

3

k

1

k

2

k

3

θ

e q u i l a t e r a l s t r e t c h e d s q u e e z e d

• 1

≤ c

• s

θ <

• .

5

• .

5 < c

• s

θ ≤ 1 B < B >

θ θ

c

• s

θ =

• .

5

+ _ Watkinson+2017

Shedding light on reionization with 21cm 37

21cm bispectra tracing ionization topology?

(2π)

3 B( ⃗

k1, ⃗ k2, ⃗ k3) δD( ⃗ k1+ ⃗ k2+ ⃗ k3)=⟨Δ( ⃗ k1) Δ( ⃗ k2) Δ( ⃗ k3) ⟩

Bispectrum is the Fourier transform of the 3-point correlation function:

contributes only when ⃗ k1+ ⃗ k2+ ⃗ k3=0

B > 0 : concentrated over-densities neutral regions B < 0 : concentrated under-densities ionized regions 21cm signal δTb

|k3| increases

Δ(⃗ k)=FT [δT b(⃗ x)]

k

1

k

1

k

2

k

2

k

3

k

3

k

1

k

2

k

3

θ

e q u i l a t e r a l s t r e t c h e d s q u e e z e d

• 1

≤ c

• s

θ <

• .

5

• .

5 < c

• s

θ ≤ 1 B < B >

θ θ

c

• s

θ =

• .

5

+ _

Shedding light on reionization with 21cm 38

21cm bispectra tracing ionization topology?

B( ⃗ k1, ⃗ k2, ⃗ k3)

√k1 k2 k 3 P(k 1) P(k 2) P(k3)

e q u i l a t e r a l s t r e t c h e d s q u e e z e d

l a r g e s c a l e s s m a l l s c a l e s

reionization progresses

d

• m

i n a t e d b y n e u t r a l r e g i

• n

s d

• m

i n a t e d b y i

• n

i z e d r e g i

• n

s

Hutter+ 2019, arXiv 1907.04342 | k

1

| = | k

2

| | k

3

| i n c r e a s e s

see also Majumdar+ 2018

Shedding light on reionization with 21cm 39

21cm bispectra tracing ionization topology?

t r a c i n g R

i

• n

c

• s

θ ≃ . 5 s e n s i t i v e t

• R

n e u t r a l

r e i

• n

i z a t i

• n

p r

• c

e e d s R

i

• n

R

n e u t r a l

peak of the size distribution of the ionized regions bispectra scale peak of the size distribution of the neutral regions

Hutter+ 2019, arXiv 1907.04342

The χHI (21cm) bispectrum undergoes 3 regimes during reionization:

➢ R > Rion & R < Rneutral ➢ R < Rion & R < Rneutral ➢ R < Rion & R > Rneutral

Shedding light on reionization with 21cm 40

21cm bispectra tracing ionization topology?

B( ⃗ k1, ⃗ k2, ⃗ k3)

√k1 k2 k 3 P(k 1) P(k 2) P(k3)

i

• n

i z e d b u b b l e s g r

• w

a b u n d a n c e

• f

n e u t r a l r e g i

• n

s r i s e s

e q u i l a t e r a l s t r e t c h e d s q u e e z e d k

i

• n

= π / R

i

• n

l a r g e s c a l e s s m a l l s c a l e s d

• m

i n a t e d b y n e u t r a l r e g i

• n

s d

• m

i n a t e d b y i

• n

i z e d r e g i

• n

s

The χHI (21cm) bispectrum

➢ traces the typical ionized

bubble size at the early stages of reionization.

➢ is sensitive to the

abundance of neutral regions near the end of reionization.

Hutter+ 2019, arXiv 1907.04342 | k

1

| = | k

2

| | k

3

| i n c r e a s e s

see also Majumdar+ 2018

Shedding light on reionization with 21cm 41

21cm bispectra tracing ionization topology!

r e i

• n

i z a t i

• n

p r

• c

e e d s R

i

• n

R

n e u t r a l

peak of the size distribution of the ionized regions bispectra scale peak of the size distribution of the neutral regions

Hutter+ 2019, arXiv 1907.04342

The more the size distribution of the ionized and neutral regions is shifted to larger scales, the longer it remains in the regime where the bispectrum switches signs at cosθ ≃ 0.5 t r a c i n g R

i

• n

c

• s

θ ≃ . 5 s e n s i t i v e t

• R

n e u t r a l

larger ionized & neutral regions

Shedding light on reionization with 21cm 42

21cm bispectra tracing ionization topology!

Shedding light on reionization with 21cm 43

21cm bispectra tracing ionization topology!

The 21cm bispectrum during reionization traces the ionization topology and difgers for difgerent ionizing escape fraction models. typical size of the ionized regions becomes larger

π/k = 4h-1 Mpc

Hutter+ 2019, arXiv 1907.04342

Shedding light on reionization with 21cm 44

Conclusions

21cm – LAE CROSS CORRELATIONS:

➢ Synergising 21cm observations with the underlying galaxy populations (specially

LAEs given their precise redshifts) will allow us to put constraints on reionization and the typical ionized bubble size around the respective galaxy population. GALACTIC PROPERTIES & TOPOLOGY OF REIONIZATION:

➢ Measuring the difgerence between the small- and large-scale 21cm power spectrum

can be used to constrain the trend of the ionizing escape fraction with galactic

• properties. A negative slope is a strong indicator for a biased distribution of the

ionizing emissivity.

➢ The 21cm bispectrum provides a valuable tracer of the ionization topology with the

change of sign tracking the typical size of the ionized regions during the earlier stages of reionization.