Intensity Mapping Baryon Acoustic Oscillation Experiments Using 21-cm, Lyman Alpha and CII (158 micron) emission

Presented at Moriond Cosmology.

By Jeff Peterson CMU, March 11, 2012

CMU Cylinder Telescope Prototype Outline

• Baryon Acoustic Oscillations • 21- cm Intensity Mapping • GBT Results • Planned 21 cm intensity mapping experiments • Lyman-alpha IM with Galex • 158 micron IM with ALMA or JCMT The History of Hydrogen Gas

1 10 redshift 100 1000

GBT LOFAR Lunar or Tianlai MWA Antarctic Array CHIME PAPER BINGO GMRT PaST/21CMA Image: Scientific American 2006 Alfalfa redshift survey

z= 0.05

Large scale structure matches optical surveys Baryon Acoustic Oscillations – Dark Energy Probe • CMB acoustic WMAP5 and other, Nolta et al (2008) oscillations: imprinted standard ruler, 400 Mly. Baryon Wiggles Detected by SDSS, 2df, WiggleZ

Eisenstein etal 2009 21 cm Intensity Mapping

• 21-cm Intensity Mapping, JP, etal (arXiv:0902.3091) • Wyithe, Loeb, Geil MNRS 383:1195 (2008) • Chang, Pen, JP, McDonald PRL 100, 091303(2008) • Battye, Davies, Weller MNRAS 355, 1339 (2004) Alfalfa 21-cm redshift survey Green Bank Telescope West Virginia

• 50 hours observation at z=1 • 10 hours at z=1.5

• DEEP2 Fields 2,4 deg x 0.5 deg fields

• 15 arcmin angular resolution at z=1 21cm – DEEP2 cross correlation

Chang, Pen, Bandura, JP, Nature 466, 463 (2010) Current program

-Attempt to detect 21 cm structure without use of optical survey

-Sweep Horizontally across Wigglz fields, rising, setting and transit

-Use CMB cross linked imaging algorithm

-Use SVD to remove modes with common Freq. structure.

-Residual rms about 40 microK, about the expected amplitude of 21 cm signal. Preliminary

21 cm <> optical cross correlation 21 cm Auto-correlation

Proposal under review for 9 feeds on GBT (and 3000 Hours) Fast 21-cm Intensity Mapping Telescope concepts • Many feeds on single dish (BINGO, GBT x 9) • 100 close-packed dishes (Tianlai?) • 2000 feeds on a cylindrical telescope (CHIME) • 1,000,000 dipole in a filled aperture 16, 3.7 meter diameter dishes and is capable of array (FFT Tel.) detecting large scale structure in HI in about a year of observing (Fig. 4).

4.1 Optimization of the HSA for Dark Energy and LSS Studies Figure 2 shows a possible configuration for the complete HSA dark energy survey telescope. Op- timization of the design is described in Ansari et al. [2], who have investigated the effect of array Figure 2: General configuration of the Hy- design on the standard ruler test with BAO, us- drogen Structure Array -Demonstrator. ing the Dark Energy Task Force Figure of Merit HSA-D is a 4 4 close-packed arrangement of × as a benchmark. They also perform simulations 3.7-m diameter dish antennas. of foreground removal assuming the foreground spectra are smooth compared to that of the HI signal. They optimize the survey parameters and needed. present a fiducial example of an HSA BAO sur- We considered four telescope configurations vey that is competitive with other Stage III dark for HSA: 1) a large single dish with a multi- energy experiments. feed array; 2) a packed set of cylinders; 3) an array of small dishes; and 4) an array of dipoles- 4.2 Experience with the Pittsburgh above-ground-plane, like PAPER or Omniscope. Cylinder Telescope (PCT) We have quite a bit of experience with option 1) through our work at GBT. The cost of a new 50 As a first step toward developing a dedicated 21- m dish at a quiet site would be prohibitive, and cm intensity mapping telescope, Peterson led an as shown in Fig. 4, the sensitivity falls short of international team including Uros Seljak, Ue-Li our goals. We also have experience with option Pen, Kris Sigurdson, and their students, to as- 2), having built a two-cylinder telescope (PCT). semble the PCT. The purpose of this prototype The cylinder choice works best when the line feed instrument was to examine cylinder telescope is fully sampled, and this would require thou- construction methods, costs, feed technologies, sands of receivers. As shown in Fig. 4, option LNAs, correlation technology, image artifacts, 3) provides excellent sensitivity with less than calibration techniques and more generally, to ex- a thousand receivers. Option 4) has the small- amine the issue of continuum foreground sub- est collecting area per receiver and this means traction. that about 10,000 receivers would be needed to The PCT consists of two cylinders each 10 m accomplish our goals. Currently the cost of the wide and 25 m long, with a separation of 25 m hardware to sample and Fourier-transform a sig- between centers. The cylinders are oriented N-S. nal is about $500 per channel, making this option PCT is a transit telescope, and uses FFT beam- unattractive in terms of cost. Dipole systems forming to create a fan of beams spanning most are also more demanding in terms of calibration of the meridian. Each cylinder currently has 32 since dipoles see all the bright sources across the dipoles installed along the focus, 16 dipoles of sky at all times. For these reasons we have cho- each linear polarization, which feeds a 64 chan- sen to construct an array of inexpensive dish an- nel software correlator. So far we have used tennas with simple feed electronics for HSA-D. dipoles for each polarization that are spaced 0.7 The HSA-D prototype instrument is formed from λ apart, so for beams near the zenith there are

5 The Tianlai Consortium

• Jeff Peterson (CMU) • Kevin Bandura, • Chen Xuelei (NAOC) • Bruce Taylor • Shi Huli • Jim McGee • Wang Yougang • Florence Liu • Wu Fangquan • Deena Kim • Li Yichao • Bruce McWilliams • Chen Zhiping (Hangzhou Dianzi University) • Ue-LI Pen (CITA) • Rui Lui

• Uros Seljak (U. C Berkeley) • Christope Yeche (Irfu - CEA) • Hee Jong Seo • Christohpe Magneville, Jim Rich • P. Abbon, C.Flouzat, H. Deschamps (~ 1 FTE) • Peter Timbie (U. Wisc.) • Reza Ansari (LAL – CNRS/IN2P3, U. Paris Sud) • Scott Dodelson (FNAL) • Jean-Eric Campagne, Marc Moniez, Anna-Sofia • John Marriner Torrento • Dave McGinnis • D. Charlet, C. Beigbeder, T. Caceres, B. • Albert Stebbins Mansoux, C. Pailler, M. Taurigna, (2.5 FTE)

• Observatoire de Paris: • Tzu-Ching Chang (IAA Taipei) • Pierre Colom, Jean-Michel Martin • Kris Sigurdson (UBC) • J. Pezzani, …

• Jon Bunton (CSIRO) Commercial satellite Dishes, individual mounts Directivity = 9.91 dB

0 dB = 9.42 dBi

Directivity = 9.91 dB

0 dB = 9.42 dBi

-40 -30 -20 -16 -13 -10-8 -6 -4 -2 Broad-band feed

-40 -30 -20 -16 -13 -10-8 -6 -4 -2

720.000 MHz 0.0 deg. 1440.000 MHz 0.0 deg. 960.000 MHz 0.0 deg. 1200.000 MHz 0.0 deg.

/Users/jeffreypeterson/Desktop/research/reionization/sky-temp/cone-dipole-both-discs-1GHz.nc Thursday 27 Oct 2011 12:01

Disk 720.000 MHz 0.0 deg. 1440.000 MHz 0.0 deg. 960.000 MHz 0.0 deg. 1200.000 MHz 0.0 deg.

/Users/jeffreypeterson/Desktop/research/reionization/sky-temp/cone-dipole-both-discs-1GHz.nc Thursday 27 Oct 2011 12:01Dipole

Ferrite Conical Reflector

180 degree hybrid LNA Noise and gain =5>?;@ABB:# 0DE9BE;6FE959># .>>EB>8K9>## 0DEEBC8KD9# @84AC59># 0GH#9D:B@I,J# LB6MDEN#OM56P?B@#I(J# QGH#LD:B@#I,J# '�?899BC@#

!"#$# ./0# 789:#$;'# QRS# !"#%# ""1# !"#$;'# 789:#(;+# !"#&# "2#3456# 789:#,;%%# !"#'# # 789:#%&;%)#

!"#(# ./0# 789:#$;'# !"#)# ""1# !"#(;+# 789:#(;+# !"#*# "2#3456# 789:#,;%%# !"#+# # 789:#%&;%)#

!" !" !" !" !" !" !" !" !" !" !" !"

!"#&,# ./0# 789:#$;'# !"#&-# ""1# !"#&,;'%# 789:#(;+# !"#'$# "2#3456# 789:#,;%%# !"#'%# # 789:#%&;%)#

/5@N## 789:#$#;#'# 789:#(##;#+### 789:#,#;%%# 789:#%&;%)# O6DE8>B# !"#$;'%# !"#$;'%# !"#$;'%# !"#$;'%#

%$#QRS# 789:###$<##%# !"#$;'%# #

789:###&<##'# !"#$;'%# #

!" !" !" !"

789:#%(<%)# !"#$;'%# # Sky Brightness Valleys

Lyman alpha C+ 158 radio Lyman Alpha Intensity Mapping with GALEX

Slit-less spectroscopy

Redshift 0.5 – 1.3 Rotate satellite to create a CAT scan of cosmic structure z 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 105 airglow FUV airglow NUV H2 Fluorescence Galex redshift range Diffuse UV Bkgr 104 ZL at (45,45) Lya w/ev Lya wo/ev

103

A s sr) 2 2 10 /cm a ( a I 101 Sunlight scattered by dust

100 Ly alpha Brightness

10-1 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 h (A) Lyman Alpha cross correlated with C+(158 micron)

Make 3-d image of SDF at 1 mm With ALMA or CSO or JCMT Subaru LAEs at z~6 Conclusions

• 21 cm intensity mapping at z~1 is succeeding • Many 21-cm BAO instruments proposed or under construction – Bingo, Tianlai, Chime, GBT nine-element etc. • Intensity mapping with Ly Alpha, 158 micron line also promising. • One year of observations with Galex could produce an IM BAO surveys