S-PASS, a new view of the polarized sky
Gianni Bernardi
On behalf of the S-PASS team
CMB2013, Okinawa, June 10-14th 2013
CMBleaks from CMB2013…
Thank you for the great conference!
Motivation 1: Polarized synchrotron emission is a foreground
for CMB observations;
Motivation 2: Galactic science (particularly the magnetic field)
Synchrotron polarized emission  magnetic field orientation (and strength)
Synchrotron polarized emission at multiple frequencies  Faraday rotation  measurement of
the magnetic field strength, orientation and direction parallel to the line of sight
Θ = Θ0
RM~220 rad m−2
0.03 cm−3
+ RMλ2
3 μG
RM =
3 kpc
ne B ⋅ d l
Motivation 2: Galactic science
37543 extragalactic RMs (reprocessing of NVSS data, Taylor et al. 2009)
simultaneously show large-scale coherence & small-scale fluctuations
Predecessor: the Parkes Galactic Meridian Survey (PGMS)
PI: E. Carretti
• 5° deg b-strip at l ~ 254°;
• 2.3 GHz, 160 MHz bandwidth (2 MHz channels
with high spectral isolation to reject RFI);
• 9’ arcmin angular resolution;
• 0.3 mK/pixel sensitivity (3’ pixel);
• 0.3 on axis instrumental pol;
• < 1% off axis instrumental pol;
Carretti et al., 2010
• Cross scans and Emerson & Grave map-making technique of data at constant
• A quadratic fit to each elevation bin removes the ground pick up effectively
• Disc region b > -20°,
small scale structure;
• Halo region b < -40°, lack
of small scale structure,
smooth emission;
Foregrounds for CMB B-mode
Evolution: the S-band Parkes All-Sky Survey (S-PASS)
E. Carretti (PI), G. Bernardi. B.M. Gaensler. M. Haverkorn, M.Kesteven, S. Poppi. L. Staveley-Smith
• To survey the polarized emission of the entire southern sky at 2.3 GHz
Dec < 0º (unshaded area);
Parkes: 2.3 GHz ;
224 MHz BW (100+ ch);
FWHM = 9’;
sbeam < 1.0 mK;
– 2000 h
– 175 nights in 2.5 yrs
– Started 07, completed in 2010
– Goals: synchrotron emission, Galactic magnetic field, CMB foregrounds
The 1.4 GHz scenario
• All sky maps at 1.4 GHz, FWHM  36’;
• Single channel surveys: no RM measures;
• Galactic Disc strongly depolarized |b| < 30°  call for higher
frequency (depolarization is frequency dependent)
Mapping: fast (120°/8 min), extended AZ scans (minimize
ground pick up and 1/f noise)
• small area basket weaving: not an option for S-PASS
– ground emission contamination (EL dependant)
– high speed requires significant overhead for short scans (10o-20o)
– short scans: mean emission on area scale is lost
• New exotic/non-standard scanning strategies has been developed
for S-PASS
– AZ scans
– Long AZ scans at South Pole EL to cover all Dec in one haul
– uses the Sky rotation to observe all RA 24 hrs.
– each day/night a zig-zag track is observed in the sky
– one zig-zag per night: accurate start timing is required
Results: total & polarized intensity
Results: Stokes Q
Results: Stokes U
Page et al. (2003) model
23 GHz
2.3 GHz
Giant Magnetized Outflows from the Centre of the Milky Way
Northern Lobe
Southern Lobe
Carretti et al., 2013
Synchrotron polarization is a unique probe of the ISM, particularly of the Galactic magnetic
field and the density of the thermal gas. It is also a relevant contaminant for CMB
polarization observations on large scales, particularly looking for the CMB B-mode;
We have completed a 2.3 GHz polarization survey of the southern sky with 9’ resolution. The
survey has SNR > 5 (9’ arcmin beam) over 96% of the observed sky;
We “stumbled upon” two giant polarized plumes connected with the Galactic Centre  they
morphologically match the γ-ray bubbles – and the extended Planck haze. The
plumes/haze/bubbles can be explained as the result of massive star formation at the Galactic
Centre – but it might not be the end of the story;
Several ongoing activities: foreground estimates in low emission regions, comparison with
the 1.4 GHz and 23 GHz data, analysis of bright regions (Gum nebula), point source RM;
We aim at a near future data release (~2 months?); people could use it to estimate foreground
contamination in the CMB data a/o include them in their foreground subtraction process (the
high latitude emission does not seem to be affected by Faraday rotation – wee next C-BASS
talk for more on this);

similar documents