Dynamic nonlocal passive scalar subgrid-scale turbulence modeling

Extensive experimental evidence highlights that scalar turbulence exhibits anomalous diffusion and stronger intermittency levels at small scales compared to in fluid turbulence. This renders the corresponding subgrid-scale dynamics modeling for a greater challenge date. We develop new large eddy simulation (LES) paradigm efficiently dynamically nonlocal LES of To this end, we formulate underlying model starting from filtered Boltzmann kinetic transport equation, where divergence fluxes emerges as fractional-order Laplacian term advection–diffusion model, coding superdiffusive nature Subsequently, robust data-driven algorithm estimation fractional (noninteger) exponent, we, on fly, calculate coefficient employing dynamic procedure. Our priori tests show our provides better agreement with ground-truth direct numerical data comparison conventional static Prandtl–Smagorinsky models. Moreover, order analyze stability assessing model's performance, carry out comprehensive posteriori tests. They unanimously illustrate considerably outperforms other existing functional models, correctly predicting backscattering phenomena and, same time, providing higher correlations small-to-large filter sizes. conclude proposed is amenable coarse very frameworks even strong anisotropies, applicable environmental applications.