CLASSY II: A technical Overview of the COS Legacy Archive Spectroscopic SurveY

TL;DR

The paper presents a detailed overview of the technical processes used to create a spectral atlas of 45 nearby star-forming galaxies from the CLASSY survey, focusing on UV spectral data reduction, coaddition, and analysis.

Contribution

It introduces new methodologies for extracting, reducing, and coadding UV spectra from HST COS data, especially for extended sources, and discusses their impact on scientific measurements.

Findings

Accurately accounting for flux calibration offsets affects stellar population properties.

Customized NUV extractions are crucial for metallicity diagnostics.

Spectral resolution variations do not significantly impact velocity measurements.

Abstract

The COS Legacy Archive Spectroscopic SurveY (CLASSY) is designed to provide the community with a spectral atlas of 45 nearby star-forming galaxies which were chosen to cover similar properties as those seen at high-z (z>6). The prime high level science product of CLASSY is accurately coadded UV spectra, ranging from ~1000-2000A, derived from a combination of archival and new data obtained with HST's Cosmic Origins Spectrograph (COS). This paper details the multi-stage technical processes of creating this prime data product, and the methodologies involved in extracting, reducing, aligning, and coadding far-ultraviolet (FUV) and near-ultraviolet (NUV) spectra. We provide guidelines on how to successfully utilize COS observations of extended sources, despite COS being optimized for point sources, and best-practice recommendations for the coaddition of UV spectra in general. Moreover, we discuss the effects of our reduction and coaddition techniques in the scientific application of the CLASSY data. In particular, we find that accurately accounting for flux calibration offsets can affect the derived properties of the stellar populations, while customized extractions of NUV spectra for extended sources are essential for correctly diagnosing the metallicity of galaxies via CIII] nebular emission. Despite changes in spectral resolution of up to ~25% between individual datasets (due to changes in the COS line spread function), no adverse affects were observed on the difference in velocity width and outflow velocities of isolated absorption lines when measured in the final combined data products, owing in-part to our signal-to-noise regime of S/N<20.