We evaluate the ability of SPIDER, a balloon-borne polarimeter, to detect a divergence-free polarization pattern (B-modes) in the Cosmic Microwave Background (CMB). In the inflationary scenario, the amplitude of this signal is proportional to that of
the primordial scalar perturbations through the tensor-to-scalar ratio r. We show that the expected level of systematic error in the SPIDER instrument is significantly below the amplitude of an interesting cosmological signal with r=0.03. We present a scanning strategy that enables us to minimize uncertainty in the reconstruction of the Stokes parameters used to characterize the CMB, while accessing a relatively wide range of angular scales. Evaluating the amplitude of the polarized Galactic emission in the SPIDER field, we conclude that the polarized emission from interstellar dust is as bright or brighter than the cosmological signal at all SPIDER frequencies (90 GHz, 150 GHz, and 280 GHz), a situation similar to that found in the Southern Hole. We show that two ~20-day flights of the SPIDER instrument can constrain the amplitude of the B-mode signal to r<0.03 (99% CL) even when foreground contamination is taken into account. In the absence of foregrounds, the same limit can be reached after one 20-day flight.
We report on our knowledge of Galactic foregrounds, as well as on how a CMB satellite mission aiming at detecting a primordial B-mode signal (CMBPol) will contribute to improving it. We review the observational and analysis techniques used to constra
in the structure of the Galactic magnetic field, whose presence is responsible for the polarization of Galactic emissions. Although our current understanding of the magnetized interstellar medium is somewhat limited, dramatic improvements in our knowledge of its properties are expected by the time CMBPol flies. Thanks to high resolution and high sensitivity instruments observing the whole sky at frequencies between 30 GHz and 850 GHz, CMBPol will not only improve this picture by observing the synchrotron emission from our galaxy, but also help constrain dust models. Polarized emission from interstellar dust indeed dominates over any other signal in CMBPols highest frequency channels. Observations at these wavelengths, combined with ground-based studies of starlight polarization, will therefore enable us to improve our understanding of dust properties and of the mechanism(s) responsible for the alignment of dust grains with the Galactic magnetic field. CMBPol will also shed new light on observations that are presently not well understood. Morphological studies of anomalous dust and synchrotron emissions will indeed constrain their natures and properties, while searching for fluctuations in the emission from heliospheric dust will test our understanding of the circumheliospheric interstellar medium. Finally, acquiring more information on the properties of extra-Galactic sources will be necessary in order to maximize the cosmological constraints extracted from CMBPols observations of CMB lensing. (abridged)
