S. Prunet et al., GALACTIC DUST POLARIZED EMISSION AT HIGH-LATITUDES AND CMB POLARIZATION, Astronomy and astrophysics (Berlin), 339(1), 1998, pp. 187-193
With recent instrumental advances, it might become possible to measure
the polarization of the Cosmic Microwave Background (CMB), e.g. by fu
ture space missions like MAP and Planck Surveyor. In this paper, we es
timate the dust polarized emission in our galaxy which is the major fo
reground to cope with for measuring the CMB polarization in the Wien p
art of CMB spectrum. We model the dust polarized emission in the galax
y using the three-dimensional HI maps of the Leiden/Dwingeloo survey a
t high galactic latitudes. We use the fact that the dust emission, for
a wide range of wavelengths, has a tight correlation with the HI emis
sion maps of this survey (Boulanger et al. 1996). Assuming the dust gr
ains to be oblate with axis Patio similar or equal to 2/3, which recen
t studies support, we determine the intrinsic dust polarized emissivit
y. The distribution of magnetic field with respect to the dust grain d
istribution is quite uncertain, we thus consider three extreme cases:
(1) The magnetic held is aligned with the major axis of the dust struc
ture, (2) the magnetic field has a random direction in the plane perpe
ndicular to the direction of major axis of the dust structure, and (3)
the magnetic field is unidirectional throughout. We further assume, a
s recent observations and theoretical analyses support, that the dust
grains align with the magnetic field independently of its strength. Th
e polarization reduction factor from misalignment of the direction of
polarization from the plane of the sky and the differential polarizati
on along a line of sight is calculated using these maps, to construct
two-dimensional maps of dust polarized emission. We calculate the angu
lar power spectrum of dust polarized emission from these maps and cast
it in variables which allow a direct comparison with the polarized co
mponent of the CMB. Our results, at frequencies similar or equal to 10
0 GHz, suggest that: (a) This foreground contamination is smaller than
the scalar-induced polarization of the CMB at rho greater than or sim
ilar to 200 while the tensor-induced polarization of CMB, which is an
order of magnitude smaller than the scalar-induced polarization, lies
below the foreground contamination level for rho greater than or simil
ar to 200, (b) the temperature-polarization cross correlation for dust
emission is more than an order of magnitude below the CMB signal for
rho greater than or equal to 200.