The Origin of Molecular Cloud Turbulence and its role on Determining the Star Formation Efficiency

Abstract

We suggest that molecular cloud (MC) turbulence is a consequence of the very process of MC formation by collisions of larger-scale flows in the diffuse atomic gas, which generate turbulence in the accumulated gas through bending- mode instabilities. Turbulence is thus maintained for as long as the accumulation process lasts ($\sim$ several Myr). Assuming that supersonic turbulence in MCs has the double role of preventing global collapse while promoting the formation of smaller-scale structures by turbulent compression (``turbulent fragmentation''), we then note the following properties: a) Turbulent fragmentation necessarily deposits progressively smaller fractions of the total mass in regions of progressively smaller sizes, because the smaller structures are subsets of the larger ones. b) The turbulent spectrum implies that smaller scales have smaller velocity differences. Therefore, below some scale, denoted $l_{\rm eq}$, the turbulent motions become subsonic. This is an energy {\it distribution} phenomenon, not a dissipative one. On this basis, we propose that the star formation efficiency (SFE) is determined by the fraction of the total mass that is deposited in clumps with masses larger than $M_{\rm J}(l_{\rm eq})$, the Jeans mass at scale $l_{\rm eq}$, because subsonic turbulence cannot promote any further subfragmentation. In this scenario, the SFE should be a monotonically increasing function of the sonic and turbulent equality scale, $l_{\rm eq}$. We present preliminary numerical tests supporting this prediction, and thus the suggestion that (one of the) relevant parameter(s) is $l_{\rm eq}$, and compare with previous proposals that the relevant parameter is the energy injection scale.