Planck Reference Sky versus WMAP

The aim of this work is to investigate the limitation and possible improvements of the Planck Reference Sky via the comparison to the 3-years WMAP data.We carry out the analysis for both temperature and polarization maps.From the Planck Reference Sky we generate the simulated maps at the five bands of WMAP, including diffuse foreground emission and compact sources.We compare the two maps in terms of power spectra and morphologically, for different sky cuts.We find discrepancies in the power spectra, particularly evident in polarization and at the lowest frequencies.Focusing on the synchrotron model we identify the possible causes of the discrepancy, and we propose improvements to the current model.We note that just dropping off the polarization fraction can give a better consistency with real data set.To be on the safe side, we compare the two data sets to an external one (DRAO-Leiden survey), confirming our results.