On average molecular clouds convert only a small fraction epsilon_ff of their mass into stars per free-fall time, but differing star formation theories make contrasting claims for how this low mean efficiency is achieved. To test these theories, we need precise measurements of both the mean value and the scatter of epsilon_ff, but high-precision measurements have been difficult because they require determining cloud volume densities, from which we can calculate free-fall times. Until recently, most density estimates assume clouds as uniform spheres, while their real structures are often filamentary and highly non-uniform, yielding systematic errors in epsilon_ff estimates and smearing real cloud-to-cloud variations. We recently developed a theoretical model to reduce this error by using column density distributions in clouds to produce more accurate volume density estimates. In this letter, we apply this model to recent observations of 12 nearby molecular clouds. Compared to earlier analyses, our method reduces the typical dispersion of epsilon_ff within individual clouds from 0.35 dex to 0.31 dex, and decreases the median value of epsilon_ff over all clouds from ~ 0.02 to ~ 0.01. However, we find no significant change in the ~ 0.2 dex cloud-to-cloud dispersion of epsilon_ff, suggesting the measured dispersions reflect real structural differences between clouds.
Download