We investigate turbulence generated by cosmological structure formation by means of large eddy simulations using adaptive mesh refinement. In contrast to the widely used implicit large eddy simulations, which resolve a limited range of length scales and treat the effect of turbulent velocity fluctuations below the grid scale solely by numerical dissipation, we apply a subgrid-scale model for the numerically unresolved fraction of the turbulence energy. For simulations with adaptive mesh refinement, we utilize a new methodology that allows us to adjust the scale-dependent energy variables in such a way that the sum of resolved and unresolved energies is globally conserved. We test our approach in simulations of randomly forced turbulence, a gravitationally bound cloud in a wind, and the Santa Barbara cluster. To treat inhomogeneous turbulence, we introduce an adaptive Kalman filtering technique that separates turbulent velocity fluctuations on resolved length scales from the non-turbulent bulk flow. From the magnitude of the fluctuating component and the subgrid-scale turbulence energy, a total turbulent velocity dispersion of several 100 km/s is obtained for the Santa Barbara cluster, while the low-density gas outside the accretion shocks is nearly devoid of turbulence. The energy flux through the turbulent cascade and the dissipation rate predicted by the subgrid-scale model correspond to dynamical time scales around 5 Gyr, independent of numerical resolution.

We extend our previous three-dimensional, full-star simulations of the final hours of convection preceding ignition in Type Ia supernovae to higher resolution using the adaptive mesh refinement capability of our low Mach number code, MAESTRO. We report the statistics of the ignition of the first flame at an effective 4.34 km resolution, and general flow field properties at an effective 2.17 km resolution. We find that off-center ignition is likely, with radius of 50 km most favored and a likely range of 40 to 75 km. This is consistent with our previous coarser (8.68 km resolution) simulations, implying that we have achieved sufficient resolution in our determination of likely ignition radii. The dynamics of the last few hot spots preceding ignition suggest that a multiple ignition scenario is not likely. With improved resolution, we can more clearly see the general flow pattern in the convective region, characterized by a strong outward plume with a lower speed recirculation. We show that the convective core is turbulent with a Kolmogorov spectrum and has a lower turbulent intensity and larger integral length scale than previously thought (on the order of 16 km s$^{-1}$ and 200 km, respectively), and we discuss the potential consequences for the first flames.

We develop a one-dimensional theoretical model for thermals burning in Type Ia supernovae based on the entrainment assumption of Morton, Taylor and Turner. Extensions of the standard model are required to account for the burning and for the expansion of the thermal due to changes in the background stratification found in the full star. The model is compared with high-resolution three-dimensional numerical simulations, both in a uniform environment, and in a full-star setting. The simulations in a uniform environment present compelling agreement with the predicted power-laws and provide model constants for the full-star model, which then provides excellent agreement with the full-star simulation. The importance of the different components in the model are compared, and are all shown to be relevant. An examination of the effect of initial conditions was then conducted using the one-dimensional model, which would have been infeasible in three dimensions. More mass was burned when the ignition kernel was larger and closer to the center of the star. The turbulent flame speed was found to be important during the early-time evolution of the thermal, but played a diminished role at later times when the evolution is dominated by the large-scale hydrodynamics responsible for entrainment. However, a higher flame speed effectively gave a larger initial ignition kernel and so resulted in more mass burned. This suggests that future studies should focus on the early-time behavior of these thermals (in particular, the transition to turbulence), and that the choice of turbulent flame speed does not play a significant role in the dynamics once the thermal has become established.

In the distributed burning regime, turbulence disrupts the internal structure of the flame, and so the idea of laminar burning propagated by conduction is no longer valid. The nature of the burning depends on the turbulent Damkohler number (Da), which steadily declines from much greater than one to less that one as the density decreases to a few 10^6 g/cc. Scaling arguments predict that the turbulent flame speed s, normalized by the turbulent intensity u, follows s/u=Da^1/2 for Da<1. The flame in this regime is a single turbulently-broadened structure that moves at a steady speed, and has a width larger than the integral scale of the turbulence. The scaling is predicted to break down at Da=1, and the flame burns as a turbulently-broadened effective unity Lewis number flame. We refer to this kind of flame as a lambda-flame. The burning becomes a collection of lambda-flames spread over a region approximately the size of the integral scale. While the total burning rate continues to have a well-defined average, s_{T} ~ u, the burning is unsteady. We present a theoretical framework, supported by both 1D and 3D numerical simulations, for the burning in these two regimes. Our results indicate that the average value of s can actually be roughly twice u for Da>1, and that localized excursions to as much as five times u can occur. The lambda-flame speed and width can be predicted based on the turbulence in the star and the turbulent nuclear burning time scale of the fuel. We propose a practical method for measuring these based on the scaling relations and small-scale computationally-inexpensive simulations. This suggests that a simple turbulent flame model can be easily constructed suitable for large-scale distributed supernovae flames.

In previous studies, we examined turbulence-flame interactions in carbon-burning thermonuclear flames in Type Ia supernovae. In this study, we consider turbulence-flame interactions in the trailing oxygen flames. The two aims of the paper are to examine the response of the inductive oxygen flame to intense levels of turbulence, and to explore the possibility of transition to detonation in the oxygen flame. Scaling arguments analogous to the carbon flames are presented and then compared against three-dimensional simulations for a range of Damk\"ohler numbers ($\Da_{16}$) at a fixed Karlovitz number. The simulations suggest that turbulence does not significantly affect the oxygen flame when $\Da_{16}<1$, and the flame burns inductively some distance behind the carbon flame. However, for $\Da_{16}>1$, turbulence enhances heat transfer and drives the propagation of a flame that is {\em narrower} than the corresponding inductive flame would be. Furthermore, burning under these conditions appears to occur as part of a combined carbon-oxygen turbulent flame with complex compound structure. The simulations do not appear to support the possibility of a transition to detonation in the oxygen flame, but do not preclude it either.

The large range of time and length scales involved in type Ia supernovae (SN Ia) requires the use of flame models. As a prelude to exploring various options for flame models, we consider, in this paper, high-resolution three-dimensional simulations of the small-scale dynamics of nuclear flames in the supernova environment in which the details of the flame structure are fully resolved. The range of densities examined, 1 to $8 \times 10^7$ g cm$^{-3}$, spans the transition from the laminar flamelet regime to the distributed burning regime where small scale turbulence disrupts the flame. The use of a low Mach number algorithm facilitates the accurate resolution of the thermal structure of the flame and the inviscid turbulent kinetic energy cascade, while implicitly incorporating kinetic energy dissipation at the grid-scale cutoff. For an assumed background of isotropic Kolmogorov turbulence with an energy characteristic of SN Ia, we find a transition density between 1 and $3 \times 10^7$ g cm$^{-3}$ where the nature of the burning changes qualitatively. By $1 \times 10^7$ g cm$^{-3}$, energy diffusion by conduction and radiation is exceeded, on the flame scale, by turbulent advection. As a result, the effective Lewis Number approaches unity. That is, the flame resembles a laminar flame, but is turbulently broadened with an effective diffusion coefficient, $D_T \sim u' l$, where $u'$ is the turbulent intensity and $l$ is the integral scale. For the larger integral scales characteristic of a real supernova, the flame structure is predicted to become complex and unsteady. Implications for a possible transition to detonation are discussed.