National Radio
Astronomy Observatory

4.4.2.1 Bandpass Normalization

The solnorm parameter (§ 4.4.1.6) deserves more explanation in the context of the bandpass. Most users are used to seeing a normalized bandpass, where the mean amplitude is unity and fiducial phase is zero. The toggle solnorm=True allows this. However, the parts of the bandpass solution normalized away will be still left in the data, and thus you should not use solnorm=True if the bandpass calibration is the end of your calibration sequence (e.g. you have already done all the gain calibration you want to). Note that setting solnorm=True will NOT rescale any previous calibration tables that the user may have supplied in gaintable.

You can safely use solnorm=True if you do the bandpass first (perhaps after a throw-away initial gain calibration) as we suggest above in § 4.2, as later gain calibration stages will deal with this remaining calibration term. This does have the benefit of isolating the overall (channel independent) gains to the following gaincal stage. It is also recommended for the case where you have multiple scans on possibly different bandpass calibrators. It may also be preferred when applying the bandpass before doing gaincal and then fluxscale (§ 4.4.4), as significant variation of bandpass among antennas could otherwise enter the gain solution and make (probably subtle) adjustments to the flux scale.

We finally note that solnorm=False at the bandpass step in the calibration chain will in the end produce the correct results. It only means that there will be a part of what we usually think of the gain calibration inside the bandpass solution, particularly if bandpass is run as the first step.

More information about CASA may be found at the CASA web page

Copyright © 2010 Associated Universities Inc., Washington, D.C.

This code is available under the terms of the GNU General Public Lincense