Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.mrao.cam.ac.uk/projects/aavp/presentations/Koopmans_EoR_SKA1.pdf Дата изменения: Thu Dec 9 18:28:02 2010 Дата индексирования: Tue Oct 2 16:41:14 2012 Кодировка: Поисковые слова: п п п п п п п

The EoR/Dark-Ages & SKA in phase 1&2
a.k.a: "The 50MHz Talk"
SKA1&2: Going from statistical detections to image data-cubes .

LИon Koopmans (Kapteyn Astronomical Institute)
Co-PI LOFAR EoR Key Science Project


General Obser vational Objectives
See also the excellent related talks by Rawlings , Greenhill, Carilli, Briggs , Pritchard & Santos
Complexity

Epoch of Reionization
LOFAR MWA PAPER

[~ 100 to ~200 MHz]

SKA1

· · · · ·

Absorption again high-z radio source (small scale power) Global signal (total power) of HI (absorption/emission) RMS fluctuations of HI emission with redshift Spatial power-spectra (1,2,3D) and higher-order statistics with redshift HI Image data-cubes (spatial-frequency) for reionization topology [~50 to ~100MHz]

Late Dark-Ages
SKA1 SKA2

· · · ·

Absorption again high-z radio source (small scale power) Global signal (total power) of HI (absorption/emission) RMS fluctuations of HI emission with redshift Spatial power-spectra (1,2,3D) and higher-order statistics with redshift

· HI Image data-cubes (spatial-frequency) for cosmology


Theoretical Motivation (I)
Total Intensity ("zero-baseline") Spatial Intensity Fluctuations (integrated over non-zero baselines)
~5
o

EoR

Late Dark Ages

EoR

Late Dark Ages

~2'

Brightness Temperature

Pritchard & Loeb 2009


Theoretical Motivation (II)
Spatial Power-Spectrum
Acore=600,000 m2 (P&L09)

SKA1: Acore=250,000 m SKA2: Acore=2.5 km
z=9

2

2

z=8

z=7

z=6

Trac & Chen 2007 Pritchard & Loeb 2009


Theoretical Motivation (III)

EoR

Late Dark Ages k=10 Mpc-1 k=1 Mpc-1 k=0.1 Mpc-1

SKA1&2:
SKA

Sufficient S/N to image HI on scales from arcmin to a degree over =50-200MHz
(tint=1000hr/BW=1MHz/Acore=600,000m2)

Pritchard & Loeb 2009


Theoretical Motivation (IVa)
Supersonic bulk flows suppresses SF to z~13 in ~5 Mpc patches: increase spin-temp fluctuations and substantially enhance brightness temp. fluctuations in the late Dark Ages Fluctuations can be off the chart!
= 150 MHz = 70 MHz = 50 MHz

z=20

k=0.1

The shaded area indicates the 1- errors per unit redshift expected for 1000 days of obser vations using a 500m dish like FAST.

Tseliakhovich & Hirata 2010; Dalal, Pen & Seljak 2010; Maio, Koopmans & Ciardi 2010


Theoretical Motivation (IVb)
Supersonic bulk flows suppresses SF to z~13 in ~5 Mpc patches: increase spin-temp fluctuations and substantially enhance brightness temp. fluctuations in the late Dark Ages

Suppression of SFR by order(s) of magnitude could delay coupling of the spin temperature to the gas via radiative processes and create cold patches (~5 Mpc scale) seen in absorption next to heated patches. This can cause much stronger Tb fluctuations around redshifts of 15-30 than thought before! Maio, Koopmans & Ciardi (2010)


General EoR/DA features that drive the technical requirements
General Features from EoR & Dark Ages: Most of reionization & late DA features occur at 66 (<200MHz): Diffuse polarized Galactic foregrounds (Tsky,I=100K - 4000K from 200MHz - 50MHz) Extragalactic (compact) sources: confusion noise, variability, polarization, etc. Ionosphere: diffraction & refraction, strongly wavelength and time dependent Instrumental Stability (e.g. beam, gains)


EoR+DA Science Goals for SKA phase 1 & 2
Transition from statistical to imaging capabilities.
LOFAR, MWA, PAPER SKA 1&2 See also talk by Steve Rawlings

SKA Phase 1: Imaging >100MHz (1) The capability of SKA1 to image ~3.0 mk fluctuations of neutral hydrogen at 3-sigma level in tint=1000 hrs at 150 MHz with BW=1 MHz, covering z~6-15. (2) Capability of SKA1 to cover all EoR & Dark Ages (z=6-30) features currently expected in the HI power-spectrum and total intensity both in emission and absorption. (3) Allow HI absorption at sub-KHz level against high-z radio sources. SKA Phase 1I: Imaging 50-100MHz (4) The capability of SKA1 to image ~0.3 mk fluctuations of neutral hydrogen at 3-sigma level in tint=1000 hrs at 150 MHz with BW=1 MHz, covering z~6-30. But imaging is feasible to z=30 (see three slides back).


Frequency Range, Bandwidth & Spectral Resolution
· Capabilities from 50 MHz (z=30) to 200 MHz (z=6) is important to fully cover
the expected redshift range over which the EoR occurs, probe the late Dark Ages and cover total intensity features in emission and absorption.

· Full instantaneous frequency coverage helps dramatically in constraining the
ionosphere, and reduce # unknowns and integration time, and (polarized) foreground removal: / ~1

· High spectral resolution (d<1 KHz) is critical for HI absorption studies using
high-z radio sources (vHI ~ km/s) and reduces bandwidth smearing and RFI


Sensitivity for Imaging Capabilities
· Sensitivity is needed on scales of k = 0.05 - 10 Mpc-1; At z>6 this translates to angular
scales of 1.5' < < 4o. (DA decreases beyond z=6.)

· Detection (~3-sigma) of Tb =3 mK fluctuations, BW=1MHz, @3' FWHM, @150MHz
in tint = 1000 hrs requires A/T>600 or Acore>250,000 m2 (SKA1) for Tsys=400K

· Imaging thus requires e.g. 10x current sensitivity of HBA collecting area of LOFAR. · FoV > 5o (per beam) is needed to sample all scales; larger FoV requires longer baselines
as well (ionospheric diffraction). A FoV ~ 20o is approximately optimal but requires outer baselines (Aouter~100,000 m2) up to 200 km to fully control the ionosphere.


UV-coverage & Spatial Resolution
· Fluctuations are on scales of several arcmin: full PS coverage requires ~1.5' resolution
or baselines of up to ~5 km in the core at 150MHz for z>6. Higher-z obser vations require shorter baselines for fixed k.

· A fully filled UV plane with a relatively uniform (TBD) # of visibilities per UV-annulus is · Instantaneous UV-coverage requires Nant> (Ѕb
2 max/Acore)=1/(2

needed to guarantee uniform sensitivity to all scales/modes and no loss of information.

) [minimal redundancy]. Tracking will create redundancy and build S/N. Needed to handle ionosphere and timevarying gains, etc.
(i) (ii) (iii) (iv) reduction ionospheric effects [i.e. bouter~[FoV/5d]x50 km] (see Koopmans 2010), reduction of confusion noise to the level of the EoR signal (requires ~200 km baselines), improvement of calibration on compact sources, cross-correlation/stacking studies.

· Outer baselines (<200 km) are needed for :


Field-of-View
· Min/Max FoV is set by several requirements:

(i) sampling sufficient volume (i.e. S/N and cosmic variance), (ii) control over the antenna beam (harder for larger FoV) (iii) sampling a large EM field being affected by the ionosphere (re/diffraction).

· A reasonable FoV @150MHz that limits cosmic

variance is FoV>5o and a maximum FoV that would have an imprint of <200 km at h=300 km (of ionosphere) is FoV<20o. Larger FoV will require single dipole antennas (>>106 elements) leading to substantial computational costs and harder to control beam shapes. But note that multi-beaming can be done.


Number of Antennas
· Minimum # of antennas for instantaneous UV coverage depends on the filling
factor of the core: Max antenna diameter then becomes Dmax = (23/2/) (A Beyond this diameter, no instantaneous UV can be achieved.

core/bmax)

.

(i) For Acore = 250,000 m2 (SKA1) Dant < 50m (Fov=3o) minimal redundancy limit, (ii) For Acore = 2.5 km2 (SKA2) Dant < 500m (Fov=0.3o) minimal redundancy limit.

· Hence, to have a good instantaneous UV coverage and 5o
one needs 7m < Dant < 30m. This requires Acore > 250,000 m2 giving perfect snapshot UV coverage already for SKA in phase I. Larger stations (up to 500m for Acore > 250,000 m2) can be allowed in phase II to have full instantaneous UV coverage , but would requires substantial multi-beaming to build a FoV larger than 5 degrees.


Requirements Table (1)
These requirements are set by science drivers and not necessarily feasibility. Main driver for SKA1 is the capability to image the EoR and obtain power-spectra of the Dark Ages: Requires 10x LOFAR HBA core area
Specification Sensitivity Core (1) Sensitivity Core (2) Tsys Acore(<5 km) Aouter(<200 km) Antenna FoV Polarization Core diameter Outer Antennas Core UV-coverage Antenna Layout Frequency/BW Spectral resolution Critical Frequency Correlator Requirement ( phase 1) A/Tsys >600 (*) Tb ~1 mK (@3'; ~1Jy); tint=1000hr, BW=1MHz (*) Tinst << 200K [v150MHz]-2.55 (sky) over 50-200 MHz >250,000 m2 (>10x LOFAR-core HBA) >100,000 m2; (Assumes FoV=20o, see below) 5o < FoV < 20o [Multi-beam for FoV<20o] (*) Full Stokes bmax=5 km; FWHMPSF of ~1.5'/3' @150/75MHz bout ~ [FoV/5o] x 50 km ~ 200 km for FoV = 20o (*) Instantaneous: Nant > (Ѕb2max/Acore) ~ 160 TBD = 50 - 200 MHz; / ~1 <1 kHz incoming data; 100kHz for science crit=100MHz tcorr = 1 sec; d <1KHz Requirements or Upg rade ( phase II) A/Tsys >6000 (*) Tb ~0.1 mK (@3'; ~1Jy); tint=1000hr, BW=1MHz (*) >2.5 km2 (>100x LOFAR-core HBA) >1.0 km2 From single to multi-beam

Option? 100-400MHz (SKAI) + 50-100MHz (SKA2) If two arrays (crit=50 & 100MHz)

(*) Reference frequency is 150 MHz (if not specified otherwise)


Requirements Table (2)
These requirements are set by science drivers and not necessarily feasibility. Main driver for SKA1 is the capability to image the EoR and obtain power-spectra of the Dark Ages: Requires e.g. 10x LOFAR HBA core area
Specification Sensitivity Core (1) Sensitivity Core (2) Tsys Acore(<5 km) Aouter(<200 km) Antenna FoV Polarization Core diameter Outer Antennas Core UV-coverage Antenna Layout Frequency/BW Spectral resolution Critical Frequency Correlator Drivers Imaging of EoR & P.S. of DA Imaging of EoR & P.S. of DA Instrument dominated above 150 MHz EoR Imaging @150MHz in tint=1000hrs Ionosphere, FG removal, calibration, confusion noise Full coverage of PS, Cosmic Variance, Ionosphere, reduction sidelobes Foreground (FG) removal Spatial power-spectrum Ionosphere, calibration, reduce confusion noise Reduce information "leakage"; Requires minimum # of stations for fixed Acore and bmax. PS sensitivity, Spatial & frequency coverage, FG, calibration Ionosphere, FGs, information criterium RFI, HI Absorption with few km/s velocity resolution Optimize collecting area and SNR (Tsky dominates below 150MHz) Reduce RFI & BW/time-smearing

(*) Reference frequency is 150 MHz (if not specified)


Conclusions

Change the DRM to allow going down to 50 MHz!
(see also talks by Greenhill, Pritchard, Santos, etc.)