21-cm cosmology with gmrt, gbt, crt - in2p3moriond.in2p3.fr/j10/transparents/peterson.pdf · 21-cm...

21-cm Cosmology with

GMRT, GBT, CRT

Presented at Moriond Cos.

By Jeff Peterson CMU

18 March 2010

CMU Cylinder Telescope Prototype

Outline

• GMRT-EoR program

• Intensity Mapping

• Baryon Acoustic Oscillations

• Green Bank Telescope Results

• Cylinder Telescope Concept

The Giant Metrewave Radio

Telescope EoR Program

“If there be light, then there is darkness”

Pythagoras, ~ BC 510

GMRT EoR TEAM

• Ue-Li Pen, Greg Paciga, Tzu-Ching

Chang, Jon Seivers (CITA),

• Jeff Peterson, Kevin Bandura, Jim

McGee, Bruce Taylor (CMU)

• Yashwant Gupta, Jayanata Roy,

Rajaram Nitanyanda (NCRA)

• Steven Myers (NRAO), Kris Sigurdson

(UBC) Chris Hirata (CIT)

Indian Giant Meterwave Radio Telescope

30 dishes @45m ea.

14 in 2 km central

core with

19,000 sq m dish

area.

We use 150 MHz +-

8 MHz

Cost: $12M in 1995

Features of GMRT-Right UV range for EoR

-Largest collecting area

-Full collecting area across sky

-Strong rejection of Cas A, Cygnus A, etc.

-Excellent Polarization Purity

-Systematic very different from dipole arrays

-If full sensitivity can be achieved,

GMRT can make detailed EoRimages.

Instrumental apparent polarization.

Ellipses stretched by factor 8

~50 times less cross pole than LOFAR/MWA

Simulated VHF sky images

Furlanetto,

et.al2003

142 MHz 144 146

148 150 152

154 156 158

Looking for 20 mK

patches on the sky

that are present only

for ~5 MHz

GMRT 150 MHz image, ca 2005

• RFI noise level 2K RMS…desired signal

20 mK

Improvements to GMRT• Built a PC-based correlator

– 8 bit sampling

– Long lags

– Programmable by mortals

• Installing new LNAs (Tsys 400K -> 100K)

• Developed RFI location algorithms and hardware

GMRT LNA

PC Array

Software corr.

RFI location

Current GMRT EoR limits

RFI in images

Extreme EoR model

Quasar ionization model

Stellar ionization model

Sensitivity possible with GMRT

Next steps for GMRT/EoR

• Continue location/elimination of RFI

sources

• Complete LNA installation

• Use drift scan strategy

• Add full-time Graduate Student (NSF

funding pending)

21 cm power spectrum vs. redshift

Pritchard & Loeb 2008

Dark Energy and structure

Inflationary physics

Reionization/dark ages

Three Dimensional 21-cm

Specific Intensity Mapping

aka

“Intensity Mapping”

-Chang, Pen, JBP, McDonald 2008-Wyithe, Loeb 2008-Sethi 2005

Alfalfa redshift survey

Large scale structure matches

optical surveys

Baryon Acoustic Oscillations – Dark

Energy Probe• Acoustic

oscillations

during ionized

era imprinted

standard ruler,

150 Mpc.

WMAP5 and other, Nolta et al (2008)

Baryon Wiggles

Detected by

SDSS and 2df

Eisenstein etal 2009

Spherical Shell in 3d

correlation space.

Steps Toward a High

Sensitivity Dark Energy Exp’t

• Study existing data (HIPASS result)

– Pen, Smith, JP, Chang (2009)

• Use existing telescopes on well studied

fields. Get 21-cm <-> Optical X-corr.

– GBT, Parkes, ATNF, Westerbork, Arecibo

• Test (and model) cylinder prototype

• Use CRT at vy quiet site to get precise

test

Green Bank Telescope

West Virginia

•DEEP2 Fields 2,4

2 deg x 0.5 deg

fields

•50 hours

observation at z=1

•10 hours at z=1.5

•15 arcmin angular

resolution at z=1

Beam

Size

RA-Long MapBeam

Size

21cm – DEEP2 cross correlation

Chang, Pen, Bandura, JP, submitted

Results from GBT

• 2 x 10^14 solar masses HI detected at

z~0.8

• Omega HI ~4.5 +- 1.0 x 10^-4

• Intensity Mapping seems productive

• Next: try this at Arecibo 700-800 MHz

• Program continues at GBT and Arecibo

• 2000 hrs would be needed to get first

BAO result

Proposed BAO instrument,

a Cylinder Radio Telescope

8192 signals

The Cylinder Radio Telescope Consortium

• Jeff Peterson (CMU)

• Kevin Bandura,

• Bruce Taylor

• Jim McGee

• Blake Conaugher

• Florence Liu

• Deena Kim

• Bruce McWilliams

• Uros Seljak (U. C Berkeley)

• Peter Timbie (U. Wisc.)

• Scott Dodelson (FNAL)

• John Marriner

• Chris Stoughton

• Hee Jong Seo

• Dave McGinnis

• Tzu-Ching Chang (IAA Taipei)

Kris Sigurdson (UBC)

• Ben Wang

• Ue-LI Pen (CITA)

• Gojko Vujanovic

• Hassane Darhmouai (AUI)

• Ahmed Legrouri

• Hassan Bourhous

• Rachid Benmouktar

• Christope Yeche (CEA)

• Christohpe Magneville

• Jim Rich

• 2 FTE engineer

• Reza Ansari (LAL)

• Marc Moniez

• 2 FTE engineer

• Jon Bunton (CSIRO)

CRT Science Goals

-Understand Dark Energy and the

acceleration of the universe, using BAO

technique.

-CRT will also allow study of

-Pulsars and Transients.

-Ionization by the first stars.

-Magnetic fields.

-Large Scale Structure P(k)

-SETI

Simulated Results

N. Gnedin R. Ansari, etal

BAO survey volumes

`CRT 2013CRT 2014

Parameters of CRT

• Aperture filling: nearly full

• Size: 100 m

• Sky coverage: 20,000 sq degrees

• Number of data streams: 64-8,000

• Freq Range: start at ~700 MHz, allow 300-

1500

• Instantaneous Bandwidth >200 MHz

CMU prototype cylinders.

Antenna and LNA

Sampling electronics, LAL,

CEA

Saclay-Irfu engineers: Ph.

Abbon,

E. Delagnes, H.

Deschamps, P. Kestener

Irfu physicists: J.-M. Le

Goff,

Ch. Magneville, J. Rich,

Ch.Yeche

Orsay-LAL engineers:Ch.

Beigbeder, D. Breton, T.

Caceres, D. Charlet,

B. Mansoux, C. Pailler, M.

Taurigna

LAL physicists: R. Ansari,

M. Moniez

Analog

electro

nics Clock distribution

500MHz

ADC board

Correlaltion computer

(Sigurdson)

Preliminary Results

Lag (20µs pixels)Lag (20µs pixels)

Tim

e (

20 s

ec p

ixels

)

Tim

e (

20 s

ec p

ixels

)

SunCassiopeia A

Next steps for CRT

• Numerically model foreground subtraction (FNAL) and compare

to data from prototype.

• Build 2-3 Cylinders 10x50 m

• Use existing 32-64 chan. Correlator

• Set up at an existing reasonably quiet observatory

– DRAO, GT, Ooty, Green Bank, Kalihari, Chihuahua

• Measure 21 cm LSS, velocity distortions

• Demonstrate successful foreground subtraction