Molecular Gas Velocity Dispersions in the Andromeda Galaxy

TL;DR

This study analyzes molecular gas velocity dispersions in Andromeda using CO emission line profiles from single-dish and interferometric observations, revealing a low brightness, extended component filtered out by interferometry.

Contribution

It identifies a broad, extended molecular gas component in Andromeda and compares velocity dispersions across different spatial scales and observational methods.

Findings

Single dish line widths are 1.5 times broader than interferometric ones.

A narrow component with FWHM ~7.5 km/s is present in both data sets.

A broad component with FWHM ~14.4 km/s is detected only in single dish data.

Abstract

In order to characterize the distribution of molecular gas in spiral galaxies, we study the line profiles of CO:1-0 emission in Andromeda, our nearest massive spiral galaxy. We compare observations performed with the IRAM 30m single-dish telescope and with the CARMA interferometer at a common resolution of 23 arcsec ~ 85pc x 350pc and 2.5 km/s. When fitting a single Gaussian component to individual spectra, the line profile of the single dish data is a factor 1.5+-0.4 larger than the interferometric data one. This ratio in line widths is surprisingly similar to the ratios previously observed in two other nearby spirals, NGC 4736 and NGC 5055, but measured at ~0.5-1 kpc spatial scale. In order to study the origin of the different line widths, we stack the individual spectra in 5 bins of increasing peak intensity and fit two Gaussian components to the stacked spectra. We find a unique narrow component of FWHM = 7.5+-0.4 km/s visible in both the single dish and the interferometric data. In addition, a broad component with FWHM = 14.4+-1.5 km/s is present in the single-dish data, but cannot be identified in the interferometric data. We interpret this additional broad line width component detected by the single dish as a low brightness molecular gas component that is extended on spatial scales >0.5 kpc, and thus filtered out by the interferometer. We search for evidence of line broadening by stellar feedback across a range of star formation rates but find no such evidence on ~100 pc spatial scale when characterizing the line profile by a single Gaussian component.