VLA pipeline - BHianca/VLA GitHub Wiki

The VLA pipeline can be found here https://science.nrao.edu/facilities/vla/data-processing/pipeline but remember this will be the most up to date version. There are always details when it comes to compatibility with versions of CASA and the standard pipeline. For this tutorial I decided to show the scripted pipeline which can be found here https://science.nrao.edu/facilities/vla/data-processing/pipeline/scripted-pipeline. I downloaded CASA_pipeline5.3.1 and if you are new to radio astronomy interferometry it might be useful to go through some background reading and look at the different steps of the standard calibration pipeline. https://science.nrao.edu/science/meetings/2019/vla-data-reduction/VLA_Pipeline.pdf

We need to modify the main file of the pipeline to include the path to the directory where it was downloaded and un-tarzipped.

Find the pwd for the pipeline and copy and paste it into the CASA terminal inside the function execfile(), such that inside the ()'s is the complete path to the pipeline (as shown below). Use single quotes at the beginning and end to make it a string.

Find the pwd for the data set that we want to run the pipeline on and copy and paste it into the CASA terminal inside the function execfile(), such that inside the ()'s is the complete path to the ./ms folder as shown below. DO NOT make the path and .ms/ a string with quotes - the pipeline already assumes as much. Then say yes to the next two prompts for real columns and Hannging smoothing on.

Now the pipeline really starts going on its own for a while - probably multiple hours.

Before this point, there was a lot of front end matter and setting up files and initializing memory. Now at this point the science data is actually starting to be calibrated to the sources chosen from the observation.

After the calibration pipeline is finished, inside the directory of the measurement set (.ms) that we reduced, you will see lots of new folders and files. Find the folder "weblog" so we can look at the general results of the calibration of the data.

Inside of the weblog folder is a bunch of plot files and tables. We can easily look at them all through a webpage. Open a new window of a browser and find the "index.html" link/file to launch a webpage.

This calibration pipeline webpage has a few headers at the top. The left most is the splash page with just some basic information. The observation summary tab has the weather and array configuration. We can look and see where each antenna by number was during the observation. Because the VLA is changed every few months, it is not given that the same antenna is in the same position for each configuration because remember there are 28 total antennas with one always off-line being worked on and re-introduced in while another is then brought off so that the number of VLA antennas in the array is always 27. The QA2 page has the calibration results. We can see from this that antenna 2 was used as the reference antenna to which all the other dishes were synched. It will have flat solutions while the others will vary slightly. This page is where if something does not look right working down the line would be the first place to look for an antenna that might have been corrupted or behaving badly during the observation. It would be easy to find here rather than looking through the raw data; first of course.