The axis of evil - a polarization perspective

We search for an unusual alignment of the preferred axes of the quadrupole and octopole, the so-called axis of evil, in the CMB temperature and polarization data from WMAP. We split the temperature and polarization maps into a part correlated with the respective other map, and a part uncorrelated with that. We then determine the preferred axes for the quadrupole and octopole in all four of those maps. If the alignment detected previously in the temperature maps were due to a preferred direction intrinsic to our Universe, we would expect it to be present in all four of our maps. In particular, the part of the polarization map which is uncorrelated with the temperature map serves as a truly independent probe of the axis of evil. We find a possible alignment in the two maps dominated by the temperature data, however, with a large uncertainty in the axes due to the high noise-level in the polarization data. In the two polarization data dominated maps, we find that the axis of the quadrupole roughly aligns with the axis of evil within our measurement precision, whereas the axis of the octopole does not. However, with our measurement uncertainty, the probability of such a scenario to happen by chance in an isotropic universe is of the order of 50 per cent. We also find that the so-called cold spot present in the CMB temperature map is even colder in the uncorrelated temperature map, although there is still a large uncertainty in the latter. Therefore, our analysis of the axis of evil, and a future analysis of the cold spot in the uncorrelated temperature data will strongly benefit from the polarization data expected from the Planck satellite.