An automated spectropolarimetric calibration and imaging pipeline designed for solar radio observations using the Murchision Widefield Array (MWA) radio telescope. It performs end-to-end calibration, flagging, and imaging with a focus on dynamic solar data, supporting both spectral and temporal flexibility in imaging products.
Solar radio data presents unique challenges due to the high variability and brightness of the Sun, as well as the need for high time-frequency resolution. The P-AIRCARS pipeline addresses these challenges by:
- Automating the calibration of interferometric data, including flux, phase, and polarization calibrations
- Supporting time-sliced and frequency-sliced imaging workflows
- Leveraging Dask for scalable parallel processing
- Providing hooks for integration with contextual data from other wavelegths for enhanced solar analysis
P-AIRCARS documentation is available at: p-aircars.readthedocs.io
P-AIRCARS is tested on Ubuntu22, Ubuntu 24, and CentOS 7 with Python 3.10. P-AIRCARS may not work in other operating system and python versions. If user wants to use P-AIRCARS in other environments, limited support is available in debugging or solving the issues. User may look at Containersed Use section in the docuement for these scenarios.
P-AIRCARS is distributed on PyPI. To use it, install it in isolated conda environment. If conda is not installed in your system, see document for Conda installation instructions.
-
Set some environment variable
export PYTHONNOUSERSITE=1 unset PYTHONPATH -
Create conda environment with python 3.10 with compaitable C/C++ libraries
conda create -n paircars_env --override-channels -c conda-forge python=3.10 gcc_linux-64=14 gxx_linux-64=14 gfortran_linux-64=14 cmake pkg-config pip conda activate paircars_envWe suggest using Mamba for fast conda installtion and environment creation.
-
Install P-AIRCARS in conda environment
pip install paircars -
Initiate necessary metadata
init-paircars-setup --initBy default, the necessary data will be saved in home directory and requires about 20 GB of disk space. We suggest using any other location with larger disk space and specify that by
--datadir </full/path/to/paircars_datadir>in the above command. -
Before running the pipeline, setup your data as following:
-- Create a <target_datadir> and put all coarse channel measurement sets of solar scan of a single observation ID (OBSID) inside it.
-- Create a <cal_datadir> and put all coarse channel measurement sets for calibrator observation of a single OBSID inside it.
-
Run P-AIRCARS pipeline
run-mwa-paircars <full path of target measurement set directory> --cal_datadir <full path of calibrator measurement set directory> --workdir <full path of work directory> --outdir <full path of output products directory>N.B.: Always provide the entire direcotry path. Short path or only directory name may cause errors. Keep target measurement sets for a single OBSID and calibrator measurement sets for a single OBSID must be kept in seperate directories. If calibrator is not present, do not provide these information.
That's all. You started P-AIRCARS pipeline for analysing your MWA solar observation 🎉.
-
To see all running P-AIRCARS jobs
show-paircars-status --show -
If P-AIRCARS is running in a local machine, see local log of any job using the
run-mwa-mwalogger --jobid <jobid>N.B.: If you are running P-AIRCARS is cluster environment, first checkout HPC Settings in the document for viewing P-AIRCARS log remotely using prefect dashboard.
-
Output products will be saved in :
<path of output products directory>
User can download and test entire P-AIRCARS pipeline using the sample dataset available in Zenodo: https://doi.org/10.5281/zenodo.18641232. Do not use this sample dataset for any publication without permission from the developer.
To view and analyse final image products, we recommend using SolarViewer.
P-AIRCARS is developed by Devojyoti Kansabanik (NCRA-TIFR, Pune, India and CPAESS-UCAR, Boulder, USA) and an incarnation of AIRCARS. Other contributors are, Surajt Mondal (NCRA-TIFR, Pune, India), Soham Dey (NCRA-TIFR, Pune, India), and Puja Majee (NCRA-TIFR, Pune, India). If you use P-AIRCARS for analysing your MWA solar observations, include the following statement in your paper, and cite the following papers:
This MWA solar observations are analysed using P-AIRCARS pipeline.
- Cite P-AIRCARS software in zenodo: [https://doi.org/10.5281/zenodo.18625477][kansbanikzenodo]
If you use observations before 2015, include this additonal statement and citation:
Flux calibration of the observations are done using the menthod described in the following paper.
P-AIRCARS name is given by Dr. Barnali Das (NCRA-TIFR, Pune, India)
This project is licensed under the MIT License.