Low Co Luminosities in Dwarf Galaxies

Abstract

We present maps of 12 CO J = 2 − 1 emission covering the entire star-forming disks of 16 nearby dwarf galaxies observed by the IRAM HERACLES survey. The data have 13 ′′ angular resolution, ∼ 250 pc at our average distance of D = 4 Mpc, and sample the galaxies by 10 − 1000 resolution elements. We apply stacking techniques to perform the first sensitive search for CO emission in dwarf galaxies outside the Local Group ranging from individual lines-of-sight, stacking over IR-bright regions of embedded star formation, and stacking over the entire galaxy. We detect 5 galaxies in CO with total CO luminosities of L CO 2−1 = 3 − 28 × 10 6 K km s −1 pc 2 . The other 11 galaxies remain undetected in CO even in the stacked images and have L CO 2−1 0.4 − 8 × 10 6 K km s −1 pc 2 . We combine our sample of dwarf galaxies with a large sample of spiral galaxies from the literature to study scaling relations of L CO with M B and metallicity. We find that dwarf galaxies with metallicities of Z ≈ 1/2 − 1/10 Z ⊙ have L CO of 2 − 4 orders of magnitude smaller than massive spiral galaxies and that their L CO per unit L B is 1 − 2 orders of magnitude smaller. A comparison with tracers of star formation (FUV and 24µm) shows that L CO per unit SFR is 1 − 2 orders of magnitude smaller in dwarf galaxies. One possible interpretation is that dwarf galaxies form stars much more efficiently, we argue that the low L CO /SFR ratio is due to the fact that the CO-to-H 2 conversion factor, α CO , changes significantly in low metallicity environments. Assuming that a constant H 2 depletion time of τ dep = 1.8 Gyr holds in dwarf galaxies (as found for a large sample of nearby spirals) implies α CO values for dwarf galaxies with Z ≈ 1/2 − 1/10 Z ⊙ that are more than one order of magnitude higher than those found in solar metallicity spiral galaxies. Such a significant increase of α CO at low metallicity is consistent with previous studies, in particular those of Local Group dwarf galaxies which model dust emission to constrain H 2 masses. Even though it is difficult to parameterize the dependence of α CO on metallicity given the currently available data the results suggest that CO is increasingly difficult to detect at lower metallicities. This has direct consequences for the detectability of star-forming galaxies at high redshift which presumably have on average sub-solar metallicity.

LIGHTING THE DARK MOLECULAR GAS: H2 AS A DIRECT TRACER

“…The black points show the molecular gas masses derived from the H 2 rotational model. The blue line shows a fit to this ratio derived from inverting the star formation law for HERACLES non-starburst galaxies (Schruba et al 2012). The molecular gas masses traced by the dust emission are denoted by the red points in the plot, and are shifted slightly in their metallicity values for clarity.…”

“…Recent studies have also shown that a CO in low-metallicity galaxies can be substantially higher than the Galactic value (e.g., Leroy et al 2011;Schruba et al 2012;Bolatto et al 2013) and hence we restricted our sample to galaxies with an oxygen abundance value of [ ]  + 12 log O H 8.4, using the average of their characteristic metallicities derived from a theoretical calibration and an empirical calibration as recommended by Moustakas et al (2010).…”

“…Since H 2 can self-shield from UV photons in regions where CO is photodissociated (Wolfire et al 2010), at low metallicity, not only is the CO abundance reduced, but as dust opacity is reduced and ionized regions become hotter and more porous, a CO,Gal can severely underestimate the molecular gas mass. Considerable effort has been invested in detecting and interpreting CO emission at metallicity 50 times lower than the solar metallicity, [ ] + 12 log O H ∼7.0, to assess the molecular gas content (Leroy et al 2011;Schruba et al 2012;Cormier et al 2014;Rémy-Ruyer et al 2014). Dust emission can be used to estimate the molecular gas mass in the ISM, assuming a constant DGR; however, it is essential to know the change in DGR with metallicity.…”

“…At the lowest metallicities, the DGR itself begins to scale nonlinearly with the metal content of the gas (Herrera-Camus et al 2012;Fisher et al 2014;Rémy-Ruyer et al 2014). Another method of inferring molecular gas content is to couple measured star formation rates with an assumed constant H 2 depletion timescale (t dep ), equivalent to a constant star formation efficiency in galaxies (Schruba et al 2012). None of these methods provides direct tracers of the H 2 reservoirs, and the relevant indirect tracers all rely on physical assumptions (e.g., constant or measurable values of a CO , DGR, or t dep ) which are unlikely to be valid in all environments.…”

Dusty star-forming galaxies at high redshift

“…Further studies have shown that indeed there is quite a bit of "CO-dark" molecular gas in these systems and other lowmetallicity regions in galaxies, and that they are not simply molecular gas-poor (e.g. McQuinn et al, 2012;Schruba et al, 2012;Blanc et al, 2013). The theoretical basis for this is that while H 2 can self-shield fairly easily, CO requires a column of dust approximately A V ∼ 1 to protect it from photo-dissociating radiation.…”