Algorithmic comparisons of decaying, isothermal, supersonic turbulence
S. Kitsionas, C. Federrath, R. Klessen, W. Schmidt, D. Price, J. Dursi, M. Gritschneder, S. Walch, R. Piontek, J. Kim, A. -K. Jappsen, P. Ciecielag, M. -M. Mac Low
Contradicting results have been reported in the literature with respect to the performance of the numerical techniques employed for the study of supersonic turbulence. We aim at characterising the performance of different particle-based and grid-based techniques on the modelling of decaying supersonic turbulence. Four different grid codes (ENZO, FLASH, TVD, ZEUS) and three different SPH codes (GADGET, PHANTOM, VINE) are compared. We additionally analysed two calculations denoted as PHANTOM A and PHANTOM B using two different implementations of artificial viscosity. Our analysis indicates that grid codes tend to be less dissipative than SPH codes, though details of the techniques used can make large differences in both cases. For example, the Morris & Monaghan viscosity implementation for SPH results in less dissipation (PHANTOM B and VINE versus GADGET and PHANTOM A). For grid codes, using a smaller diffusion parameter leads to less dissipation, but results in a larger bottleneck effect (our ENZO versus FLASH runs). As a general result, we find that by using a similar number of resolution elements N for each spatial direction means that all codes (both grid-based and particle-based) show encouraging similarity of all statistical quantities for isotropic supersonic turbulence on spatial scales k<N/32 (all scales resolved by more than 32 grid cells), while scales smaller than that are significantly affected by the specific implementation of the algorithm for solving the equations of hydrodynamics. At comparable numerical resolution, the SPH runs were on average about ten times more computationally intensive than the grid runs, although with variations of up to a factor of ten between the different SPH runs and between the different grid runs. (abridged)
In a systematic study, we compare the density statistics in high resolution numerical experiments of supersonic isothermal turbulence, driven by the usually adopted solenoidal (divergence-free) forcing and by compressive (curl-free) forcing. We find that for the same rms Mach number, compressive forcing produces much stronger density enhancements and larger voids compared to solenoidal forcing. Consequently, the Fourier spectra of density fluctuations are significantly steeper. This result is confirmed using the Delta-variance analysis, which yields power-law exponents beta~3.4 for compressive…
We here analyse numerical simulations of supersonic, hypersonic and magnetohydrodynamic turbulence that is free to decay. Our goals are to understand the dynamics of the decay and the characteristic properties of the shock waves produced. This will be useful for interpretation of observations of both motions in molecular clouds and sources of non-thermal radiation. We find that decaying hypersonic turbulence possesses an exponential tail of fast shocks and an exponential decay in time, i.e. the number of shocks is proportional to t exp (-ktv) for shock velocity jump v and mean initial wavenumb…
We present results of large-scale three-dimensional simulations of supersonic Euler turbulence with the piecewise parabolic method and multiple grid resolutions up to 2048^3 points. Our numerical experiments describe non-magnetized driven turbulent flows with an isothermal equation of state and an rms Mach number of 6. We discuss numerical resolution issues and demonstrate convergence, in a statistical sense, of the inertial range dynamics in simulations on grids larger than 512^3 points. The simulations allowed us to measure the absolute velocity scaling exponents for the first time. The iner…
We present the results of three-dimensional simulations of supersonic Euler turbulence with grid resolutions up to 1024^3 points. Our numerical experiments describe nonmagnetized driven turbulent flows with an isothermal equation of state and an rms Mach number of 6. We demonstrate that the inertial range scaling properties of turbulence in this strongly compressible regime deviate substantially from a Kolmogorov-like behavior previously recovered for mildly compressible transonic flows.
We investigate the hydrodynamics of three-dimensional classical Bondi-Hoyle accretion. A totally absorbing sphere of different sizes (1, 0.1 and 0.02 accretion radii) moves at different Mach numbers (0.6, 1.4, 3.0 and 10) relative to a homogeneous and slightly perturbed medium, which is taken to be an ideal, nearly isothermal, gas ($γ=1.01$). The hydrodynamics is modeled by the ``Piecewise Parabolic Method'' (PPM). The resolution in the vicinity of the accretor is increased by multiply nesting several $32^3$-zone grids around the sphere, each finer grid being a factor of two smaller in zone si…
We study the probability distribution function (PDF) of the mass density in simulations of supersonic turbulence with properties appropriate for molecular clouds. For this study we use Athena, a new higher-order Godunov code. We find there are surprisingly similar relationships between the mean of the time-averaged PDF and the turbulent Mach number for driven hydrodynamic and strong-field MHD turbulence. There is, however, a large scatter about these relations, indicating a high level of temporal and spatial variability in the PDF. Thus, the PDF of the mass density is unlikely to be a good mea…
We employ a generalization of the She & Leveque model to study velocity scaling relations based on our simulations of thermal instability--induced turbulence. Being a by-product of the interstellar phase transition, such multiphase turbulence tends to be more intermittent than compressible isothermal turbulence. Due to radiative cooling, which promotes nonlinear instabilities in supersonic flows, the Hausdorff dimension of the most singular dissipative structures, D, can be as high as 2.3, while in supersonic isothermal turbulence D is limited by shock dissipation to D < 2. We also show that s…
A roughly constant temperature over a wide range of densities is maintained in molecular clouds through radiative heating and cooling. An isothermal equation of state is therefore frequently employed in molecular cloud simulations. However, the dynamical processes in molecular clouds include shock waves, expansion waves, cooling induced collapse and baroclinic vorticity, all incompatible with the assumption of a purely isothermal flow. Here, we incorporate an energy equation including all the important heating and cooling rates and a simple chemical network into simulations of three-dimensiona…
The question of the shape of the density PDF for supersonic turbulence is addressed, using both analytical and numerical methods. For isothermal supersonic turbulence, the PDF is Log-Normal, with a width that scales approximately linearly with the Mach number. For a polytropic equation of state, with an effective gamma smaller than one, the PDF becomes skewed and reminiscent of (but not equal to) a power law on the high density side.
We performed a series of three-dimensional numerical simulations of supersonic homogeneous Euler turbulence with adaptive mesh refinement (AMR) and effective grid resolution up to 1024^3 zones. Our experiments describe non-magnetized driven supersonic turbulent flows with an isothermal equation of state. Mesh refinement on shocks and shear is implemented to cover dynamically important structures with the highest resolution subgrids and calibrated to match the turbulence statistics obtained from the equivalent uniform grid simulations. We found that at a level of resolution slightly below 512^3…
The probability density function (PDF) of the gas density in turbulent supersonic flows is investigated with high-resolution numerical simulations. In a systematic study, we compare the density statistics of compressible turbulence driven by the usually adopted solenoidal forcing (divergence-free) and by compressive forcing (curl-free). Our results are in agreement with studies using solenoidal forcing. However, compressive forcing yields a significantly broader density distribution with standard deviation ~3 times larger at the same rms Mach number. The standard deviation-Mach number relation…
Supersonic turbulence generates distributions of shock waves. Here, we analyse the shock waves in three-dimensional numerical simulations of uniformly driven supersonic turbulence, with and without magnetohydrodynamics and self-gravity. We can identify the nature of the turbulence by measuring the distribution of the shock strengths. We find that uniformly driven turbulence possesses a power law distribution of fast shocks with the number of shocks inversely proportional to the square root of the shock jump speed. A tail of high speed shocks steeper than Gaussian results from the random superp…
Molecular clouds have broad linewidths suggesting turbulent supersonic motions in the clouds. These motions are usually invoked to explain why molecular clouds take much longer than a free-fall time to form stars. It has classically been thought that supersonic hydrodynamical turbulence would dissipate its energy quickly, but that the introduction of strong magnetic fields could maintain these motions. In a previous paper it has been shown, however, that isothermal, compressible, MHD and hydrodynamical turbulence decay at virtually the same rate, requiring that constant driving occur to mainta…
Supersonic turbulence fragments the interstellar medium into dense sheets, filaments, cores and large low density voids. The turbulence is driven on large scales, probably predominantly by supernovae. The scaling properties of supersonic turbulence are well described by a new analytical theory. The distribution of proto-stellar core masses depends primarily on the power spectrum of the turbulent flow, and on the jump conditions for near isothermal shocks in a magnetized gas. For the predicted velocity power spectrum index beta=1.74, consistent with results of numerical experiments of supersoni…
We review how supersonic turbulence can both prevent and promote the collapse of molecular clouds into stars. First we show that decaying turbulence cannot significantly delay collapse under conditions typical of molecular clouds, regardless of magnetic field strength so long as the fields are not supporting the cloud magnetohydrostatically. Then we review possible drivers and examine simulations of driven supersonic and trans Alfvénic turbulence, finally including the effects of self-gravity. Our preliminary results show that, although turbulence can support regions against gravitational coll…
We performed numerical simulations of supersonic isothermal turbulence driven by mostly compressive large-scale forcing, using both a static grid and adaptive mesh refinement with an effective resolution N=768^3. After a transient phase dominated by shocks, turbulence evolves into a steady state with an RMS Mach number about 2.5, in which cloud-like structures of over-dense gas are surrounded by highly rarefied gas. The index of the turbulence energy spectrum function beta = 2.0 in the shock-dominated phase. As the flow approaches statistical equilibrium, the spectrum flattens, with beta = 1.9…
Turbulence is essential for understanding the structure and dynamics of molecular clouds and star-forming regions. There is a need for adequate tools to describe and characterize the properties of turbulent flows. One-point probability distribution functions (pdf's) of dynamical variables have been suggested as appropriate statistical measures and applied to several observed molecular clouds. However, the interpretation of these data requires comparison with numerical simulations. To address this issue, SPH simulations of driven and decaying, supersonic, turbulent flows with and without self-g…
In this study we investigate the formation and properties of prestellar and protostellar cores using hydrodynamic, self-gravitating Adaptive Mesh Refinement simulations, comparing the cases where turbulence is continually driven and where it is allowed to decay. We model observations of these cores in the C$^{18}$O$(2\to 1)$, NH$_3(1,1)$, and N$_2$H$^+(1\to 0)$ lines, and from the simulated observations we measure the linewidths of individual cores, the linewidths of the surrounding gas, and the motions of the cores relative to one another. Some of these distributions are significantly differe…
We present the results from three dimensional hydrodynamical simulations of decaying high-speed turbulence in dense molecular clouds. We compare our results, which include a detailed cooling function, molecular hydrogen chemistry and a limited C and O chemistry, to those previously obtained for decaying isothermal turbulence. After an initial phase of shock formation, power-law decay regimes are uncovered, as in the isothermal case. We find that the turbulence decays faster than in the isothermal case because the average Mach number remains higher, due to the radiative cooling. The total therm…
We present numerical studies of compressible, decaying turbulence, with and without magnetic fields, with initial rms Alfven and Mach numbers ranging up to five, and apply the results to the question of the support of star-forming interstellar clouds of molecular gas. We find that, in 1D, magnetized turbulence actually decays faster than unmagnetized turbulence. In all the regimes that we have studied 3D turbulence-super-Alfvenic, supersonic, sub-Alfvenic, and subsonic-the kinetic energy decays as (t-t0)^(-x), with 0.85 < x < 1.2. We compared results from two entirely different algorithms in t…
Colliding hypersonic flows play a decisive role in many astrophysical objects. In this paper, we look at the idealized model of a 2D plane parallel isothermal slab (CDL) and at symmetric settings, where both flows have equal parameters. We performed a set of high-resolution simulations with upwind Mach numbers, 5 < M_u < 90. We find that the CDL is irregularly shaped and has a patchy and filamentary interior. The size of these structures increases with l_cdl, the extension of the CDL. On average, but not at each moment, the solution is about self-similar and depends only on M_u. We find the ro…
