Pipeline originally developed by Russ Taylor in 2011
Modified by Ishwara Chandra in 2018
This is an improved version of the polarization pipeline by Silpa Sasikumar
Major improvements in flagging and self-calibration
Kindly refer to the paper: Ishwara-Chandra et al 2020
NOT tested for uGMRT band-2 (150 MHz)
Queries: jbaghel@ncra.tifr.res.in
Initial phase only calibration added by Silpa Sasikumar in 2019-2020
Polarization steps added by Silpa Sasikumar in 2019-2020, which involve:
(1) Flagging of polarized and unpolarized calibrators
(2) Polarization calibration (cross-hand delays; leakage terms; R-L polarization angle)
(3) Stokes 'Q' and 'U' imaging
The current version of the pipeline also flags all the four correlations (RR, LL, RL, LR).
Pipeline modifications by Janhavi Baghel in 2020-2021, changes made:
(1)refantmode = 'strict' and only one reference antenna to be specified.
(2)datacoulmn = 'corrected' in the flagging of target field before splitting
(3)datacoulmn = 'data' in tclean during creation of dirty image. Split target file has no datacolumn 'corrected'.
(4)datacoulmn = 'RESIDUAL_DATA' when flagging residuals in self-calibration cycles
(5)Initial flagging of known bad antenna from observer log.
(6)Polarization modelling included within the pipeline for 3C286,3C48,3C138 with the pol_*.txt files.
TEST.FITS is the output of gvfits, which coverts GMRT LTA format to multi-source FITS
Please CHANGE channels and source fields as per your data.
Also change clip parameters if you have much stronger calibrator and/or target
The parameters below are typical for 550 MHz, 2048 channels at Band-4 (550-750 MHz)
Please change as required for your data.
In BAND-4, recommended channels corresponding to ~ 560 MHz to ~ 810 MHz. The sensitivity drops sharply after 810 MHz.
In any case DO NOT use beyond 820 MHz.
It is highly recommended not to use OQ208 (unpolarized calibrator) to calculate the instrumental leakage for uGMRT since it is a very faint source (~few Jy); therefore a single short scan does not provide sufficient SNR to accurately determine the instrumental polarization.
Also, we do not recommend the use of 3C138 (polarized calibrator) for leakage calibration.
We recommend 3C286 (polarized calibrator) or 3C84 (unpolarized calibrator) for leakage calibration.