An uncertainty principle for star formation – V. The influence of dust extinction on star formation rate tracer lifetimes and the inferred molecular cloud lifecycle

Abstract

Recent observational studies aiming to quantify the molecular cloud lifecycle require the use of known “reference time-scales” to turn the relative durations of different phases of the star formation process into absolute time-scales. We previously constrained the characteristic emission time-scales of different star formation rate (SFR) tracers, as a function of the SFR surface density and metallicity. However, we omitted the effects of dust extinction. Here, we extend our suite of SFR tracer emission time-scales by accounting for extinction, using synthetic emission maps of a high-resolution hydrodynamical simulation of an isolated, Milky-Way-like disc galaxy. The stellar feedback included in the simulation is inefficient compared to observations, implying that it represents a limiting case in which the duration of embedded star formation (and the corresponding effect of extinction) is overestimated. Across our experiments, we find that extinction mostly decreases the SFR tracer emission time-scale, changing the time-scales by factors of 0.04–1.74, depending on the gas column density. UV filters are more strongly affected than Hα filters. We provide the limiting correction factors as a function of the gas column density and flux sensitivity limit for a wide variety of SFR tracers. Applying these factors to observational characterisations of the molecular cloud lifecycle produces changes that broadly fall within the quoted uncertainties, except at high kpc-scale gas surface densities ($\Sigma _{\rm g}\gtrsim20 {\mathrm{M_{\odot }\, pc^{-2}}}$). Under those conditions, correcting for extinction may decrease the measured molecular cloud lifetimes and feedback time-scales, which further strengthens previous conclusions that molecular clouds live for a dynamical time and are dispersed by early, pre-supernova feedback.

On the duration of the embedded phase of star formation

“…We refer the reader to Kruijssen & Longmore (2014) for a detailed explanation of the method, to Kruijssen et al (2018) for the presentation and validation of the HEISENBERG code, as well as the full list of input parameters, and to Chevance et al (2020c) for a general application of the method to nine nearby starforming galaxies. The accuracy of the method has been demonstrated in Kruijssen et al (2018) using simulated galaxies, and has since been confirmed through extensive observational and numerical testing (Kruijssen et al 2019;Haydon et al 2020a;Ward et al 2020b).…”

“…The absolute duration of the different phases is then obtained by scaling the relative duration of timescales with a reference time-scale (t ref ). In our previous analyses using CO and H α observations (Kruijssen et al 2019;Chevance et al 2020c, a;Hygate 2020;Ward et al 2020a), we used the duration of the isolated H α emitting phase (t ref = t H α − t fb, H α ), calibrated by Haydon et al (2020b), Haydon et al (2020a) using the stellar population synthesis model SLUG2 (da Silva, Fumagalli & Krumholz 2012Krumholz , 2014Krumholz et al 2015), as the reference timescale. Here, in order to obtain absolute values when applying our analysis to CO and 24 μm maps, we first apply the method to CO and H α observations.…”

H-alpha emission in local galaxies: star formation, time variability and the diffuse ionized gas

“…3). Haydon et al (2020a) revisited results from Haydon et al (2020b) by quantify the impact of dust extinction, find that extinction mostly decreases the SFR tracer emission timescale (factor of a few for H𝛼). Observationally, correlating maps of different light tracers such as H𝛼, CO and 24𝜇m on spatially resolved scales to empirically constrain the giant molecular cloud lifecycle (see also Kruijssen et al 2019;Chevance et al 2020), Kim et al (2021) found a duration of H𝛼 emitting phase of 5.5-8.9 Myr for nearby spiral galaxies, which is consistent with our dust-corrected timescale estimates.…”

Pre-supernova feedback mechanisms drive the destruction of molecular clouds in nearby star-forming disc galaxies

“…We note that the choice of the CO-to-H 2 conversion factor does not affect the time-scales determined here, unless there are considerable variations within the galaxy on the scale of independent regions (a few 100 pc; see also discussion inKruijssen et al 2018;Chevance et al 2020b). In addition, in the galaxies of our sample, ranging from solar to half-solar metallicity, we do not expect high-mass star formation to take place in completely CO-dark clouds, which would require a different calibration of the timeline(Haydon et al 2020a). …”

“…Kruijssen et al 2019;Chevance et al 2020b;Hygate 2020;Ward et al 2020;Zabel et al 2020) and are dispersed within a few Myr after the onset of high-mass star formation, as visible in H𝛼 emission (e.g. Whitmore et al 2014;Hollyhead et al 2015;Grasha et al 2018Grasha et al , 2019Hannon et al 2019;Chevance et al 2020b,a;Haydon et al 2020a). It is an open question which physical mechanisms drive GMC dispersal.…”