HETDEX: AGN-galaxy clustering at 2

HETDEX:AGN-galaxy clustering at 2<z<3.5

Andreas Schulze

Kavli Institute for Mathematics and Physics of the Universe (Kavli IPMU)The University of Tokyo

Workshop "Clustering Measurements of Active Galactic Nuclei"Garching, Germany, 14-18 July 2014

Hobby-Eberly Telescope Dark Energy Experiment

⇒ blind, large-scale spectroscopic survey with VIRUS at HET⇒ 300 square degrees (60 square degrees with spectra) over 3 years

⇒ about 0.75 million redshifts from 1.9 < z < 3.5 (Lyα emitters)⇒ about 1 million redshifts from 0 < z < 0.5 (OII emitters)

⇒ Timeline: 2015-2017

Andreas Schulze (Kavli IPMU) HETDEX 2 / 16

Hobby-Eberly Telescope (HET)

McDonald Observatory, Texasfixed elevation-axis designsegmented primary mirror11.1 x 9.8 m (9.2 m effectiveaperture)

HETDEX⇒ HET upgraded to 22 arcmin

field of view⇒ in progress

⇒ new instrument VIRUS⇒ construction complete byend of 2014


VIRUS instrument

Visible Integral-Field ReplicableUnit Spectrograph (VIRUS)

75 IFUs feeding 150spectrographseach IFU contains 448 fibers(33 600 spectra per shot)3500− 5500 Å at R∼750Line flux limit:3.5× 10−17 erg/cm2/sand mAB ∼ 22


Survey Design

⇒ Spring field:42x7 deg2 with1/4.5 fill factorcentered on13hr, +53deg

⇒ Fall field:28x5 deg2

centered on1.5hr, 0deg

⇒ Fall field with ancillary data in optical (grizy, DES), near-IR (K,NEWFIRM), mid-IR (Spitzer-SHELA), far-IR (Herschel-HeRS)


Constraining Dark Energy with HETDEX

⇒ measure power spectrum of ∼ 800,000 LAEs

⇒ measure H(z = 2.3) and DA(z = 2.3) to 0.9%⇒ constrain evolution of dark energy (wa)⇒ most precise measurements at z > 2


AGN in HETDEX

fully blind IFU spectroscopic survey data from HETDEX enablesmultitude of ancillary science, including AGN

⇒ no preselection required⇒ detection through characteristic emission lines in spectra

type-2 AGN hard to identify([Ne V]3426 line, X-ray, mid-IR, ...)type-1 (broad line) AGN easy to distinguish from LAEs and [OII]emitters⇒ several lines detectable (MgII, CIII], CIV, Lyα) up to z = 3.5⇒ unambiguous redshift identification at z > 1.3


simulated type 1 AGN spectra

⇒ z = 1

4000 4500 5000 5500wavelength λ

400

600

800

1000

counts

g=21

4000 4500 5000 5500wavelength λ

200

300

400

500

600

counts

g=22

4000 4500 5000 5500wavelength λ

100

200

300

400

counts

g=23

⇒ z = 2

4000 4500 5000 5500wavelength λ

500

1000

1500

counts

4000 4500 5000 5500wavelength λ

100

200

300

400

500

600

700

counts

4000 4500 5000 5500wavelength λ

0

100

200

300

400

counts

⇒ z = 3

4000 4500 5000 5500wavelength λ

500

1000

1500

2000

counts

4000 4500 5000 5500wavelength λ

200

400

600

800

1000

counts

4000 4500 5000 5500wavelength λ

0

100

200

300

400

500

600

counts


HETDEX AGN selection

16 18 20 22 24g−band magnitude

100

101

102

103

104

cum

ula

tive

num

berN

(<g)

allz<1.01.0<z<1.9z>1.9

predicted numbercountsmain field: ∼ 20 000 AGN> 10 000 AGN at z > 1no preselection

Predicted HETDEX AGNselection function

21 22 23 24 25

g magnitude (AB)

1

2

3

redsh

iftz


AGN demographics

⇒ excellent sample for statistical studies of type 1 AGN

AGN luminosity function and evolution from 0 < z < 3.5

Evolution of active BH mass function and Eddington ratiodistribution function

AGN-Galaxy clustering


AGN Luminosity function

-8

-7

-6

-5

log

Φ[M

pc−

3m

ag−

1]

z=0.49 z=0.87 z=1.25

-30-25-20MB [mag]

-8

-7

-6

-5

log

Φ[M

pc−

3m

ag−

1]

z=1.63

-30-25-20MB [mag]

z=2.01

-30-25-20MB [mag]

z=2.40

-30-25-20MB [mag]

-8

-7

-6

-5

log

Φ[M

pc−

3m

ag−

1]

z=2.80

-30-25-20MB [mag]

z=3.25

predicted AGN LF for HETDEX mainsurvey (SDSS, VVDS, HETDEX)

AGN Luminosity functionconstrain faint end of AGNLF between z = 2 andz = 3.5 down to g ∼ 24,without preselection


Black hole mass and Eddington ratio distributionfunction

⇒ MBH estimates frombroad lines up toz = 2.5

⇒ determine AGNevolution as functionof MBH and Lbol/LEdd

⇒ probe BH massdownsizing between0 < z < 2.5

-28-26-24-22-20

Mi(z=2) [mag]

-10

-8

-6

-4

log

Φ

7 8 9 10

log M • [M¯]

-8

-6

-4

log

Φ•

-2 -1 0

log λ

-8

-6

-4

log

Φλ

z=1.6z=0.0

0.5 1.0 1.5 2.0 2.5z

-9

-8

-7

-6

-5

-4lo

gΦ

log LBol

44.044.545.045.546.046.547.0

0.5 1.0 1.5 2.0 2.5z

log M •

7.07.58.08.59.09.510.0

0.5 1.0 1.5 2.0 2.5z

log λ

−2.0−1.7−1.4−1.1−0.8−0.5−0.2

Schulze+(in prep.)based on SDSS+VVDS+zCOSMOS



⇒ within 1.9 < z < 3.5 HETDEX will probe 104 AGN and 106

galaxies (LAEs) in the same volume

⇒ excellent sample for AGN-galaxy clustering at z > 2⇒ construct AGN-LAE 2-point cross-correlation function

(100× increase in power over AGN auto-correlation function)

Ross et al. (2009)



large scale clustering:⇒ AGN bias factor⇒ relation to DM halos (typical DM halo masses)

small scale clustering:⇒ reach below 1/h Mpc separation => nonlinear regime⇒ Halo occupation distribution (HOD) => constrain AGN fueling and

feedback

good statistics to bin in AGN luminosity, BH mass,. . .

⇒ HETDEX will provide the precise probe of AGN-galaxy clusteringat z > 2


AGN-Galaxy clustering: The MHalo −MBH relation

MHalo from clusteringMBH from spectra

⇒ MHalo −MBH relation at2 < z < 2.5

Current estimate at 1.5 < z < 3based on small samples

Adelberger & Steidel (2005)


Conclusions

HETDEX will start first half 20153 year, massive, blind spectroscopic survey in two fields over300+150 deg2

powerful AGN survey (20000 type 1 AGN in main field)AGN demographics: AGNLF, BH mass and Eddington ratiodistribution function, AGN-galaxy clusteringwithin 1.9 < z < 3.5 104 AGN and 106 galaxies in the samevolume

⇒ HETDEX will provide unique probe of AGN-galaxy clustering atz > 2


Documents

HETDEX: AGN-galaxy clustering at 2