[Previous] | [Session 69] | [Next]
M.-M. Mac Low (AMNH/MPIA), F. Heitsch (MPIA), R. S. Klessen (Leiden)
Star-forming molecular clouds appear to have lifetimes over an order of magnitude longer than the time it would take for them to freely collapse if they were supported only by thermal pressure. In previous work we have shown that, in the absence of driving, both hydrodynamical and MHD turbulence decay too quickly to explain this non-thermal support. Here we explore whether uniformly driven turbulence can support clouds against collapse. In the hydrodynamical case, we compare 3D computations performed with an Eulerian code (ZEUS-3D) at resolutions of up to 2563 to computations with an SPH code using up to several hundred thousand particles; we then compare hydrodynamical to MHD runs with ZEUS-3D.
We test the claim made by Bonnazzola et al. (1987) and Léorat et al. (1990) that turbulence can only support a region if it is driven on a scale less than the local Jeans length. This claim appears correct for support against global collapse. However, supersonic turbulence produces strong density fluctuations that have much shorter local Jeans lengths. These fluctuations will probably collapse unless the turbulence is being driven on these very short scales. In real molecular clouds, ambipolar diffusion may act to suppress driving on small scales, potentially allowing local collapse to occur even in globally supported clouds. The difference between local collapse in turbulently supported regions and global collapse in regions lacking turbulent support may correspond to the difference between isolated and clustered modes of star formation.
Computations were performed at the Rechenzentrum Garching of the MPG, at the National Center for Supercomputing Applications, and at the Hayden Planetarium.
If the author provided an email address or URL for general inquiries, it is a
s follows:
http://www.mpia-hd.mpg.de/theory/maclow/