Shock finding on a moving mesh – I. Shock statistics in non-radiative cosmological simulations

2015 
Cosmological shock waves play an important role in hierarchical structure formation by dissipating and thermalizing kinetic energy of gas flows, thereby heating the universe. Furthermore, identifying shocks in hydrodynamical simulations and measuring their Mach number accurately is critical for calculating the production of non-thermal particle components through diffusive shock acceleration. However, shocks are often significantly broadened in numerical simulations, making it challenging to implement an accurate shock finder. We here introduce a refined methodology for detecting shocks in the moving-mesh code AREPO, and show that results for shock statistics can be sensitive to implementation details. We put special emphasis on filtering against spurious shock detections due to tangential discontinuities and contacts. Both of them are omnipresent in cosmological simulations, for example in the form of shear-induced Kelvin-Helmholtz instabilities and cold fronts. As an initial application of our new implementation, we analyse shock statistics in non-radiative cosmological simulations of dark matter and baryons. We find that the bulk of energy dissipation at redshift zero occurs in shocks with Mach numbers around ${\cal M}\approx2.7$. Furthermore, almost $40\%$ of the thermalization is contributed by shocks in the warm hot intergalactic medium (WHIM), whereas $\approx60\%$ occurs in clusters, groups and smaller halos. Compared to previous studies, these findings revise the characterization of the most important shocks towards higher Mach numbers and lower density structures. Our results also suggest that regions with densities above and below $\delta_b=100$ should be roughly equally important for the energetics of cosmic ray acceleration through large-scale structure shocks.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    100
    References
    66
    Citations
    NaN
    KQI
    []