[Previous] | [Session 42] | [Next]
E. Vazquez-Semadeni, V. Avila-Reese (Instituto de Astronomia, UNAM)
We explore the dissipative ability of numerical simulations of compressible MHD turbulence that represent the interstellar medium at intermediate-to-large scales. Turbulent kinetic energy Ek is injected realistically into the medium by means of randomly placed ``stars'', which radially accelerate their surrounding medium. We characterize the input sources by their size lf and by the rate \dot{E}i,s at which each one injects energy into the flow. A third important parameter is \dot{\Sigma}OB, the rate of formation of ``OB stars''. The spatially-scattered, small-scale nature of the injection (``forcing'') gives rise to the coexistence of both forced and decaying turbulent regimes within the same flow.
In the forced regime, the global dissipation rate is always very similar to the injection rate, implying that most of the energy is dissipated in the vicinity of the sources. The characteristic dissipation time is given approximately by ti\ge \langle\Sigma_g\rangle u_{rms}^2/E_{i,s}\dot{\Sigma}_{OB}, where \langle\Sigmag\rangle is the average gas surface density, urms is the rms velocity dispersion, and Ei,s is the net kinetic energy input per source. Since urms also depends on the other injection parameters, it turns out that the dissipation time is nearly independent of \dot{E}i,s and of \dot{\Sigma}OB, the dominant parameter being lf. Empirically, we find that ti \propto lf0.7. For realistic values of the parameters, ti\approx 15-25 Myr.
In the decaying regime, the kinetic energy decays with time as ~t-0.8. This time dependence can be translated into a distance dependence. We discuss the implications of our results for models of galaxy formation and evolution.
This research has received partial funding from Conacyt grant 27752-E.
The author(s) of this abstract have provided an email address for comments about the abstract: e.vazquez@astrosmo.unam.mx