sdgaincal computes and removes a time-dependent gain variation in single-dish data on a per-spectral-window and per-antenna basis. Presently the task operates only on data taken with the ALMA fast-mapped, double-circle observation modes [1] . This task exploits the fact that the double-circle mode observes the same position in the center of the mapped field, approximately circular every sub-cycle, and normalizes the gains throughout the entire dataset, relative to the measured brightness at the center position.

Info: This gain calibration task is done independently of, and following, the atmosphere (i.e. $T_{sys}$) and sky calibration steps applied through the sdcal task. Alternative way to apply these caltables is to utilize pre-application capability of sdgaincal task. This can be done by feeding caltables into the task using applytable parameter. You can specify spw mapping and interpolation method via spwmap and interp, respectively. Usage for these parameters are exactly same as applycal.

Configurable inputs control the calibration mode, selection parameters, and output behavior:

Presently, this task has only one calibration mode: calmode='doublecircle'.

The size of the region that CASA regards as "the center" is user-configurable via the expandable 'radius' (in arcsec) parameter (under 'calmode').  The default is to use the size of the primary beam. The data can also be smoothed in the time domain, prior to computation of the gain variation. Selection is by specral window/channels, field IDs, and antenna through the spw, field, and antenna selection parameters. The default is to use all data for the gain calibration. The caltable can be output with the 'outfile' parameter.


Citation Number 1
Citation Text Phillips et al, 2015. Fast Single-Dish Scans of the Sun Using ALMA. PDF