next up previous contents index
Next: Do I really an Up: BASIC CONCEPTS OF CALIBRATION Previous: Stokes Parameters and Polarization

Should I calibrate in AIPS or MIRIAD ?

   

When you calibrate in AIPS, it is simply assumed that the calibrator is unpolarized. Calibrator sources are typically linearly polarized at a few percent, and generally not more than 5 percent. When you calibrate ATCA data with AIPS, this error goes directly into the gains that you determine. However, on application of these gains to some target source, the errors cancel out (as they are opposite in sign and equal in strength for the X and Y feeds). However, because AIPS is poorly attuned to linear feeds, you will be unable to extract any Stokes parameter other than I from your ATCA data. Finally, you will not be able to account for the antenna polarizations either (although this is a second order effect in total intensity).

When you calibrate in MIRIAD, it is possible to account for both the calibrators polarization and the antenna polarizations so that an optimum calibration is possible. However, despite the calibration virtues of MIRIAD, you should be aware that most spectral-line observations record only the XX and YY polarizations. Without the XY and YX correlations we are unable to make a full and correct polarimetric calibration, and the calibration in MIRIAD would be the same as the one in AIPS.

In general, to get the best possible calibration of any XX, YY, XY and YX observation, whether you are interested in total intensity only or all Stokes parameters, you should calibrate in MIRIAD as described in the MIRIAD User's Guide. If you intend to calibrate in MIRIAD, then you need to concern yourself with the XY phase differences which are discussed below in § 3.5. If you have only XX and YY polarizations, both AIPS and MIRIAD make the same calibration.


next up previous contents index
Next: Do I really an Up: BASIC CONCEPTS OF CALIBRATION Previous: Stokes Parameters and Polarization

nkilleen@atnf.csiro.au