Increased molecular gas velocity dispersion and star formation efficiency in barred galaxy centres

TL;DR

The paper investigates how stellar bars influence central molecular gas and star formation by analyzing high-resolution PHANGS-ALMA CO$(2-1)$ data with a linewidth-dependent $\alpha_{\rm{CO}}$. Using central hexagonal pixels to compare five properties, it finds that barred centers exhibit higher velocity dispersion $\sigma_{\rm{v}}$, higher intensity-weighted molecular surface density $\Sigma_{\rm{mol,IW}}$, and shorter gas depletion times $t_{\rm dep}$, with some differences reduced when AGN are removed. Linear-regression analyses show that $\sigma_{\rm{v}}$ correlates with $\Sigma_{\rm{mol,IW}}$ and that the Kennicutt–Schmidt relation in the centers has a slope near unity for both barred and unbarred galaxies, though barred centers sit at higher $\sigma_{\rm{v}}$ for a given gas content. The study suggests bar-driven inflow induces extra non-circular motions and turbulence, lowering $t_{\rm dep}$ and potentially enhancing star formation in bars’ central regions, while emphasizing that results depend on the $\alpha_{\rm{CO}}$ prescription and on the availability of high-resolution data.

Abstract

Work by the Physics at High Angular resolution in Nearby GalaxieS (PHANGS) collaboration found higher molecular gas surface densities and velocity dispersions in the centres of barred galaxies compared to unbarred galaxies. We explore central molecular gas using published high resolution (150 pc) measurements of CO$(2-1)$ from the PHANGS-ALMA survey and a new velocity dispersion-dependent prescription for the CO-to-H$_{2}$ conversion factor $α_{\rm{CO}}$. Comparisons of the molecular gas surface density, velocity dispersion, star formation rate, and depletion time reveal that these quantities are different in the centres of barred and unbarred galaxies. Gas depletion times are found to be shorter in barred galaxy centres. Even when we control for the presence of an AGN, the velocity dispersion and depletion time are found to be statistically different between barred and unbarred galaxy centres. The higher velocity dispersion suggests extra non-circular motions, possibly due to the inflow of gas along the bar, that are not constant but must increase as the star formation rate increases.

Figures (5)

• Figure 1: Boxplots comparing the distributions of properties of the central regions for the full sample of barred (42) and unbarred (22) galaxies for each property [top row] and for a subset of the sample containing only galaxies with no AGN, leaving barred (30) and unbarred (19) [bottom row]. From left to right, the plots show: area-averaged molecular gas surface density; intensity-weighted molecular gas surface density; intensity-weighted velocity dispersion; SFR surface density; and gas depletion time (both area-averaged). The box encloses the region between the first and third quartiles with the median line shown in orange. Whiskers extend out to 1.5 times the interquartile range and black circles indicate outliers. The gas calculations are done using the linewidth-based $\alpha_{\rm{CO}}$ from Teng2024. We find moderate to strong differences between the two full samples for $\Sigma_{\rm{mol,IW}}$, $\sigma_{\rm{v}}$, $\Sigma_{\rm{SFR}}$, and $t_{\rm{dep}}$. However, when galaxies with an AGN are removed from the sample, the only significant differences between the two distributions are found for $\sigma_{\rm{v}}$ and $t_{\rm{dep}}$.
• Figure 2: Barred versus unbarred resolved galaxy properties in the central region. [Top] $\sigma_{\rm{v}}$ as a function of $\Sigma_{\rm{mol,IW}}$ for the full sample, and [bottom] $\sigma_{\rm{v}}$ as a function of $\Sigma_{\rm{mol,IW}}$ for the sample with galaxies with an AGN removed. Best fit lines generated with Linmix with slope and intercept uncertainties shown in shaded regions. Even once galaxies with an AGN are removed, barred galaxies still show increased $\sigma_{\rm{v}}$ in the central region.
• Figure 3: [Left] The Kennicutt-Schmidt relation $\Sigma_{\rm{SFR}}$ as a function of $\Sigma_{\rm{mol,RA}}$ for barred versus unbarred resolved molecular gas observations of the central region. The two samples agree very well. [Right] $\sigma_{\rm{v}}$ as a function of $\Sigma_{\rm{SFR}}$ for barred versus unbarred resolved central galaxy properties. Barred galaxies show somewhat increased $\sigma_{\rm{v}}$ in the central region. Best fit lines generated with Linmix with slope and intercept uncertainties shown in shaded regions.
• Figure 4: Boxplots comparing the distributions of the full sample of barred (42) and unbarred (22) galaxies [top row] and sample with no AGN [bottom row] for each property, using the metallicity dependent $\alpha_{\rm{CO}}$. See Figure \ref{['fig:boxResolvedT24']} for description of the quantities and method. The most significant differences between barred and unbarred galaxies for both the full sample and the sample with no AGN, are found for $\sigma_{\rm{v}}$ and $\Sigma_{\rm{mol,IW}}$ (See Table \ref{['tab:ADwholeSample']}).
• Figure 5: Barred versus unbarred resolved galaxy properties in the central region using the metallicity-based $\alpha_{\rm{CO}}$ from SunJiayi2022multi [Top] $\sigma_{\rm{v}}$ as a function of $\Sigma_{\rm{mol,IW}}$ for the full sample; [Bottom] $\sigma_{\rm{v}}$ as a function of $\Sigma_{\rm{mol,IW}}$ for galaxies without an AGN. Best fit lines generated with Linmix with slope and intercept uncertainties are shown in shaded regions. Barred galaxies without AGN still show increased $\sigma_{\rm{v}}$ in the central region.