The life cycle of the Central Molecular Zone – I. Inflow, star formation, and winds

Abstract

We present a study of the gas cycle and star formation history in the central 500 pc of the Milky Way, known as Central Molecular Zone (CMZ). Through hydrodynamical simulations of the inner 4.5 kpc of our Galaxy, we follow the gas cycle in a completely self-consistent way, starting from gas radial inflow due to the Galactic bar, the channelling of this gas into a dense, star-forming ring/stream at ≈ 200 − 300 pc from the Galactic centre, and the launching of galactic outflows powered by stellar feedback. We find that star formation activity in the CMZ goes through oscillatory burst/quench cycles, with a period of tens to hundreds of Myr, characterised by roughly constant gas mass but order-of-magnitude level variations in the star formation rate. Comparison with the observed present-day star formation rate of the CMZ suggests that we are currently near a minimum of this cycle. Stellar feedback drives a mainly two-phase wind off the Galactic disc. The warm phase dominates the mass flux, and carries 100 − 200% of the gas mass converted into stars. However, most of this gas goes into a fountain and falls back onto the disc rather than escaping the Galaxy. The hot phase carries most of the energy, with a time-averaged energy outflow rate of 10 − 20% of the supernova energy budget.

Simulations of the Milky Way’s Central Molecular Zone – II. Star formation

“…It is therefore natural to ask: What drives the time variability in the SFR of the CMZ? A possible explanation is that the CMZ goes through episodic starbursts driven by feedback instabilities (Krumholz & Kruijssen 2015;Krumholz, Kruijssen & Crocker 2017;Torrey et al 2017;Armillotta et al 2019). In this scenario, the CMZ has a roughly constant gas mass but order-of-magnitude level variations in the SFR.…”

“…The large scatter (∼1 dex) in the depletion times observed in the centre of external barred galaxies (Leroy et al 2013;Utomo et al 2017) is explained by temporal fluctuations. Armillotta et al (2019) run numerical hydrodynamical simulations of gas flowing in a barred potential that included star formation prescriptions that lend support to this scenario. In their simulation, the CMZ depletion time is not 2, which is a good indication of shear for a 2D flow.…”

“…Second, the star formation prescription used in Armillotta et al (2019) is also quite different from that used in our code. In their approach star particles are stochastically formed in gas denser than 10 3 cm −3 , provided that it is gravitationally bound, cold, and self-shielded.…”

“…Compared to our scheme, the main differences are their choice of density threshold and the fact that in their scheme, significant quantities of dense gas can accumulate above the density threshold, something that is impossible by design in our scheme. Third, Armillotta et al (2019) include the effects of photoionization feedback and SN feedback, while we concentrate here solely on the latter.…”

“…The CMZ has emerged in the last decade as a unique laboratory for the study of star formation (e.g. Molinari et al 2011;Kruijssen et al 2014;Armillotta et al 2019). The main reason is that the environmental conditions in which stars are born are more extreme than anywhere else in the Galaxy.…”