[Paper Review] PypeIt: The Python Spectroscopic Data Reduction Pipeline

J. X. Prochaska, Joseph F. Hennawi+6 more|arXiv (Cornell University)|Apr 7, 2020

Astronomy and Astrophysical Research9 references123 citations

TL;DR

PypeIt is a Python package for semi-automated reduction of astronomical spectroscopic data, supporting multiple instruments and providing comprehensive online documentation and parameters.

ABSTRACT

Major version release of the PypeIt software package.

Motivation & Objective

• Enable semi-automated reduction of diverse astronomical spectroscopic data.
• Provide a unified workflow with documented input parameters and functionality.
• Support a broad set of spectrographs through reproducible, open-source tooling.
• Leverage decades of development in prior pipelines to streamline data processing.

Proposed method

• Implement a Python-based reduction pipeline with modular components.
• Document the full input parameter space and pipeline functionality online.
• Provide instrument-specific support for listed spectrographs (e.g., Gemini/GMOS, Keck/DEIMOS, VLT/X-Shooter, etc.).
• Base algorithms on established spectroscopic reduction techniques from prior pipelines.
• Offer release versions with tested instrument compatibility (e.g., v1.0.3).
• Collaborate with the community and plan for submission to JOSS.

Experimental results

Research questions

• RQ1 How can we unify spectroscopic data reduction across multiple instruments in a single Python package?
• RQ2 What is the scope of instruments supported by PypeIt in its initial v1.0.3 release?
• RQ3 How can semi-automated reduction improve reproducibility and accessibility of spectroscopic data?
• RQ4 What documentation and online resources best communicate the pipeline's parameters and functionality?

Key findings

• PypeIt provides semi-automated data reduction for a wide range of instruments.
• The release v1.0.3 supports numerous spectrographs including Gemini/GNIRS, Gemini/GMOS, Keck/DEIMOS, Keck/MOSFIRE, VLT/X-Shooter, and others.
• The reduction procedure and input parameter list are documented online and regularly updated.
• The project builds on decades of prior pipeline development to inform its design.
• The authors indicate plans to submit the work to the Journal of Open Source Software (JOSS).