The goal of this code is to be able to run on an a set of ALMA or VLA *_targets.ms (or other MS files with the targets split out and having the same setup in each MS) files for single-pointing data and perform self-calibration using the continuum. If a cont.dat file is present, the code will flag the non-continuum regions and perform self-calibration on the continuum only.
To run this code with a concatenated calibrated_final.ms that one might receive from the NA ARC, one must split out the groups of SPWs associated with the individual observations, selecting on SPW, such that one has a set of MSes with SPWs that are all the same. For example, if an MS has SPWs 0,1,2,3,10,11,12,13, and 0,1,2,3 are from the first observation and 10,11,12,13 are from the second observation, they should be split out as follows:
split(vis='my_concat.ms',spw'0,1,2,3',outputvis='my_ms_0_targets.ms')
split(vis='my_concat.ms',spw'10,11,12,13',outputvis='my_ms_1_targets.ms')This code can only be executed within the CASA 6.4+ environment, and is tested on CASA 6.4.0, 6.4.1, 6.4.2, 6.4.3, 6.4.4, and 6.5.0. CASA versions earlier that 6.4.0 are likely to encounter problems due to a gaincal bug that was fixed in CASA 6.4.0.
Standard ALMA and single-band VLA data are supported as are multiple targets. Multi-band VLA is supported. ALMA spectral scans may work, but have not been thoroughly tested. Mosaics have only had limited testing, but some limited functionality might work, specifically small mosaics where there is emission of the same source(s) within each pointing. Imaging will use nterms=2 for most VLA data and nterms=1 for most ALMA data. However, on a per-target basis, nterms=2 imaging will be used if S/N > 500, otherwise the full improvement possible from self-calibration may not be realized. Amplitude self-calibration will be attempted after phase-only self-calibration. If amplitude self-cal is not desired, set the variable 'do_amp_selfcal' to False in the auto_selfcal.py script.
-
Create an empty directory
-
Copy into this directory the
*_targets.msfiles that have identical setups (targets and spectral windows) to be self-calibrated (must contain only the targets desired for self-calibration and only TARGET observation intents) -
Copy auto_selfcal.py and selfcal_helpers.py into this directory (and regenerate_weblog.py if needed)
-
Run script with
mpicasa -n X casa -c auto_selfcal.py # X is the number of mpi threads to use.If serial operation is desired (without mpicasa), run with,
casa -c auto_selfcal.py
Alternatively, one could try it out using the modular casa6 packages in a custom Conda environment (see casa6.yaml)
xvfb-run -d python auto_selfcal.py
The script will automatically find your *_targets.ms files, and determine all the parameters it needs for imaging and self-calibration of the science targets.
The script will output a file applycal_to_orig_MSes.py to apply the self-calibration solutions back to the original MS such that the line data can also have the self-calibration solutions applied.
The script will output a file uvcontsub_orig_MSes.py if a cont.dat file exists, excluding the same spectral regions for continuum fitting as were flagged for continuum self-calibration.
When finished summaries are generated for each target in weblog/index.html. Weblog generation may fail depending on what settings are cached for the viewer. If this is a problem you can remove all your cached casaviewer settings by running rm -rf ~/.casa/viewer. Then you can run casa -c regenerate_weblog.py, the regenerate_weblog.py script is included in the repository and only generates a new weblog from the already stored images and metadata pickle files.
-
Certain functions to convert from LSRK to channel, S/N estimates, and tclean wrapper have their origins from the ALMA DSHARP large program reduction scripts.
-
The functions to parse the cont.dat file and convert to channel ranges (used the routine from above) was adapted from a function written by Patrick Sheehan @psheehan for the ALMA eDisk large program