S. Kitsionas, R. Klessen, C. Federrath, W. Schmidt, D.J. Price, L.J. Dursi, M. Gritschneder, R. Piontek, Jongsoo Kim, A.-K. Jappsen, P. Ciecielag and M.-M. Mac Low

[ pdf (1.2Mb) ] Accepted to A&A, 24th Sep 2009

---

Abstract

Context. Simulations of astrophysical turbulence have reached a level of sophistication that quantitative results are now starting to emerge. Contradicting results have been reported, however, in the literature with respect to the performance of the numerical techniques employed for its study and their relevance to the physical systems modelled.
Aims. We aim at characterising the performance of a number of hydrodynamics codes on the modelling of turbulence decay. This is the first such large-scale comparison ever conducted.
Methods. We have driven compressible, supersonic, isothermal turbulence with GADGET and then let it decay in the absence of gravity, using a number of grid (ENZO, FLASH, TVD, ZEUS) and SPH codes (GADGET, VINE, PHANTOM).
Results. We have analysed the results of our numerical experiments using a variety of statistical measures ranging from energy spectrum functions (power spectra), to velocity structure functions, to probability distribution functions.
Conclusions. In the low numerical resolution employed here the performance of the various codes is comparable. In more detail, our analysis indicates that the numerical techniques used can be sorted from least to most dissipative as follows: ENZO-FLASH; TVD; ZEUS-SPH codes. Use of the Morris & Monaghan viscosity implementation for SPH results in less dissipation. We have shown that the density-weighted power spectrum is a more robust statistical measure for the study of compressible turbulence. Here we have adopted the (rho/rho_0 )^1/2 velocity weights which provide physical reference to kinetic energy.

Movie

Movie of the PHANTOM calculation (6Mb Quicktime)

Images

Snapshots from the PHANTOM calculation: