Calibration at P Band is particularly troublesome because of confusion in the primary beam. Confusing sources contribute several Jy or more to every field of view, and the use of narrow bandwidths for interference rejection implies that these sources contribute significantly to virtually all baselines. External calibrators must therefore be chosen for their strength rather than by nearness to the interesting field of view. Calibrators with >25 Jy are preferable. Unless the calibrator is >5 Jy, you may as well attempt self-calibration using the brightest sources in the field. Self-calibration with the traditional algorithms may not work under severe ionospheric conditions, where the size of the isoplanatic patch may become smaller than that of the primary beam. Perley (1988) has listed possible P Band calibrators.
Because ionospheric Faraday rotation at P Band is large and variable, polarization calibration is extremely difficult and cannot be guaranteed unless the ionosphere is exceptionally ``quiet" during an observing run.
Note also that at P Band, only the D configuration satisfies the conditions for imaging using a two-dimensional Fourier transform(Lecture 10) with a non-coplanar array. P Band observations in the more extended configurations must either use special-purpose software (consult Tim Cornwell) to correct for the effects of the non-coplanar baselines, or accept more approximate removal of only the strongest confusing sources in the field using MX or IMAGR in AIPS.