Dr. Mark Krumholz
Princeton University

University of Florida Astronomy Colloquium - Nov 9th, 2005

Turbulence, Feedback, and the Global Evolution of Giant Molecular Clouds

Star formation is extremely inefficient, with giant molecular clouds converting only ~1% of their gas into stars per dynamical time. Explaining this result, known since the 1970s, remains one of the outstanding problems in the theory of star formation. I present a model, called feedback-driven turbulence, to solve the puzzle. In feedback-driven turbulence, star formation is inefficient because turbulent motions prevent a majority of the gas in GMCs from collapsing. In turn, the star formation that does occur drives turbulence at a rate that offsets the loss of energy by radiation from isothermal shocks. This balance between energy loss and driving keeps clouds in virial balance and determines the star formation rate. The model is able to reproduce from first principles a number of observational results, including the total star formation rate in the Milky Way, the Kennicutt-Schmidt Law, Larson's Law of constant GMC column densities, and the age spreads of OB associations. The model also makes new quantitative predictions that will be testable in the next few years.