Search for a command to run...
We use 3D numerical MHD simulations to follow the evolution of cold, turbulent, gaseous systems with parameters representing GMC conditions. We study three cloud simulations with varying mean magnetic fields, but identical initial velocity fields. We show that turbulent energy is reduced by a factor two after 0.4-0.8 flow crossing times (2-4 Myr), and that the magnetically supercritical cloud models collapse after ~6 Myr, while the subcritical cloud does not collapse. We compare density, velocity, and magnetic field structure in three sets of snapshots with matched Mach numbers. The volume and column densities are both log-normally distributed, with mean volume density a factor 3-6 times the unperturbed value, but mean column density only a factor 1.1-1.4 times the unperturbed value. We use a binning algorithm to investigate the dependence of kinetic quantities on spatial scale for regions of column density contrast (ROCs). The average velocity dispersion for the ROCs is only weakly correlated with scale, similar to the mean size-linewidth relation for clumps within GMCs. ROCs are often superpositions of spatially unconnected regions that cannot easily be separated using velocity information; the same difficulty may affect observed GMC clumps. We analyze magnetic field structure, and show that in the high density regime, total magnetic field strengths increase with density with logarithmic slope 1/3 -2/3. Mean line-of-sight magnetic field strengths vary widely across a projected cloud, and do not correlate with column density. We compute simulated interstellar polarization maps at varying orientations, and determine that the Chandrasekhar-Fermi formula multiplied by a factor ~0.5 yields a good estimate of the plane-of sky magnetic field strength provided the dispersion in polarization angles is < 25 degrees.