VISIONS: the VISTA Star Formation Atlas

II. The data processing pipeline^★

¹ Universität Wien, Institut für Astrophysik, Türkenschanzstrasse 17, 1180 Wien, Austria
e-mail: stefan.meingast@univie.ac.at
² Laboratoire d’Astrophysique de Bordeaux, Univ. Bordeaux, CNRS, B18N, Allée Geoffroy Saint-Hillaire, 33615 Pessac, France
³ Max-Planck-Institut für Astronomie, Königstuhl 17, 69117, Heidelberg, Germany

Received: 22 December 2022
Accepted: 28 February 2023

Abstract

The VISIONS public survey provides large-scale, multi-epoch imaging of five nearby star-forming regions at sub-arcsecond resolution in the near-infrared. All data collected within the program and provided by the European Southern Observatory (ESO) science archive are processed with a custom end-to-end pipeline infrastructure to provide science-ready images and source catalogs. The data reduction environment has been specifically developed for the purpose of mitigating several shortcomings of the bona fide data products processed with software provided by the Cambridge Astronomical Survey Unit (CASU), such as spatially variable astrometric and photometric biases of up to 100 mas and 0.1 mag, respectively. At the same time, the resolution of co-added images is up to 20% higher compared to the same products from the CASU processing environment. Most pipeline modules are written in Python and make extensive use of C extension libraries for numeric computations, thereby simultaneously providing accessibility, robustness, and high performance. The astrometric calibration is performed relative to the Gaia reference frame, and fluxes are calibrated with respect to the source magnitudes provided in the Two Micron All Sky Survey (2MASS). For bright sources, absolute astrometric errors are typically on the order of 10–15 mas and fluxes are determined with sub-percent precision. Moreover, the calibration with respect to 2MASS photometry is largely free of color terms. The pipeline produces data that are compliant with the ESO Phase 3 regulations and furthermore provides curated source catalogs that are structured similarly to those provided by the 2MASS survey.