Accuracy

The goal should be to map Stokes V, Q, and U limited by thermal noise and not by instrumental effects. As a practical matter, the goal should be instrumental polarization effects of <0.1%, after calibration. Moreover, this spec must be met over the entire primary beams of the telescopes in order to map over the single primary beam and to mosaic map. A significant difference between standard intensity (Stokes I) mapping and polarization mapping is instrumental polarization. For intensity mapping, the primary beam is a relatively simple and stable function, so the instrumental response (dirty beam) can be predicted from the UV coverage. Knowledge of that instrumental response can therefore be used to deconvolve it out of the final maps. The instrumental response in Stokes parameters Q, U, and V depends in addition to the UV coverage on the polarized instrumental response over the primary beams of the various antennas, and in general this may vary strongly and in a complicated manner with position in the primary beam, time, pointing position, etc. In order to deconvolve the polarized dirty beams out of the final polarization maps, the polarized dirty beams must be known at the noise level of the maps. If the instrumental polarization due to the antennas is stable in time, one can measure it once and take it out. Time variable instrumental polarization (due to elevation effects for example) requires great loss of sensitivity due to time spent on calibration and/or limitations on polarization fidelity. Failure to know the polarized response of the instrument over position and time is the major limitation on the accuracy of polarization mapping.