CASA on the fringe -- Development of VLBI processing capabilities for CASA

TL;DR

This paper details the development of VLBI data processing capabilities in CASA, enabling more accessible, reproducible, and modern workflows for VLBI observations, with results comparable or superior to traditional software.

Contribution

Introduction of VLBI processing tasks in CASA, enhancing its functionality for VLBI data calibration and analysis, and demonstrating its effectiveness against AIPS.

Findings

CASA can process VLBI data with results matching AIPS.

CASA's VLBI processing is more accessible and suitable for pipelines.

CASA performs better in some cases, despite longer processing times.

Abstract

New functionality to process Very Long Baseline Interferometry (VLBI) data has been implemented in the CASA package. This includes two new tasks to handle fringe fitting and VLBI-specific amplitude calibration steps. Existing tasks have been adjusted to handle VLBI visibility data and calibration meta-data properly. With these updates, it is now possible to process VLBI continuum and spectral line observations in CASA. This article describes the development and implementation, and presents an outline for the workflow when calibrating European VLBI Network or Very Long Baseline Array data in CASA. Though the CASA VLBI functionality has already been vetted extensively as part of the Event Horizon Telescope data processing, in this paper we compare results for the same dataset processed in CASA and AIPS. We find identical results for the two packages and conclude that CASA in some cases performs better, though it cannot match AIPS for single-core processing time. The new functionality in CASA allows for easy development of pipelines or Jupyter notebooks, and thus contributes to raising VLBI data processing to present day standards for accessibility, reproducibility, and reusability.