Dependence of the star formation efficiency on global parameters of molecular clouds

Abstract

We investigate the response of the star formation efficiency (SFE) to the main parameters of simulations of molecular cloud formation by the collision of warm diffuse medium (WNM) cylindrical streams, neglecting stellar feedback and magnetic fields. The parameters we vary are the Mach number of the inflow velocity of the streams, Msinf, the rms Mach number of the initial background turbulence in the WNM, and the total mass contained in the colliding gas streams, Minf. Because the SFE is a function of time, we define two estimators for it, the "absolute" SFE, measured at t = 25 Myr into the simulation's evolution (sfeabs), and the "relative" SFE, measured 5 Myr after the onset of star formation in each simulation (sferel). The latter is close to the "star formation rate per free-fall time" for gas at n = 100 cm^-3. We find that both estimators decrease with increasing Minf, although by no more than a factor of 2 as Msinf increases from 1.25 to 3.5. Increasing levels of background turbulence similarly reduce the SFE, because the turbulence disrupts the coherence of the colliding streams, fragmenting the cloud, and producing small-scale clumps scattered through the numerical box, which have low SFEs. Finally, the SFE is very sensitive to the mass of the inflows, with sferel decreasing from ~0.4 to ~0.04 as the the virial parameter in the colliding streams increases from ~0.15 to ~1.5. This trend is in partial agreement with the prediction by Krumholz & McKee (2005), since the latter lies within the same range as the observed efficiencies, but with a significantly shallower slope. We conclude that the observed variability of the SFE is a highly sensitive function of the parameters of the cloud formation process, and may be the cause of significant scatter in observational determinations.