AutoSpec: Fast Automated Spectral Extraction Software for IFU Datacubes

TL;DR

AutoSpec is a new Python software that automates the extraction of high-quality spectra from large IFU datacubes by utilizing spectral information for source identification, improving signal-to-noise, and deblending sources without relying on ancillary imaging.

Contribution

The paper introduces AutoSpec, a novel automated spectral extraction tool that leverages spectral data within IFU datacubes, enhancing extraction quality and efficiency over traditional methods.

Findings

Comparable to deep photometry weighted extractions

Improves signal-to-noise ratio of spectra

Successfully deblends sources from nearby contaminants

Abstract

With the ever growing popularity of integral field unit (IFU) spectroscopy, countless observations are being performed over multiple object systems such as blank fields and galaxy clusters. With this, an increasing amount of time is being spent extracting one dimensional object spectra from large three dimensional datacubes. However, a great deal of information available within these datacubes is overlooked in favor of photometrically based spatial information. Here we present a novel, yet simple approach of optimal source identification, utilizing the wealth of information available within an IFU datacube, rather than relying on ancillary imaging. Through the application of these techniques, we show that we are able to obtain object spectra comparable to deep photometry weighted extractions without the need for ancillary imaging. Further, implementing our custom designed algorithms can improve the signal-to-noise of extracted spectra and successfully deblend sources from nearby contaminants. This will be a critical tool for future IFU observations of blank and deep fields, especially over large areas where automation is necessary. We implement these techniques into the Python based spectral extraction software, AutoSpec which is available via GitHub at: https://github.com/a-griffiths/AutoSpec and Zenodo at: https://doi.org/10.5281/zenodo.1305848