Abstract:We consider a drift-diffusion process with a time-independent and divergence-free random drift that is of white-noise character. We are interested in the critical case of two space dimensions, where one has to impose a small-scale cut-off for well-posedness, and is interested in the marginally super-diffusive behavior on large scales. In the presence of an (artificial) large-scale cut-off at scale L, as a consequence of standard stochastic homogenization theory, there exist harmonic coordinates with a stationary gradient $F_L$; the merit of these coordinates being that under their lens, the drift-diffusion process turns into a martingale. It has recently been established that the second moments diverge as $\mathbb{E}|F_L|^2\sim\sqrt{\ln L}$ for $L\uparrow\infty$. We quantitatively show that in this limit, and in the regime of small Péclet number, $|F_L|^2/\mathbb{E}|F_L|^2$ is not equi-integrable, and that $\mathbb{E}|{\rm det}F_L|/\mathbb{E}|F_L|^2 $ is small. Hence the Jacobian matrix of the harmonic coordinates is very peaked and non-conformal. We establish this asymptotic behavior by characterizing a proxy $\tilde F_L$ introduced in previous work as the solution of an Itô SDE w. r. t. the variable $\ln L$, and which implements the concept of a scale-by-scale homogenization based on a variance decomposition and admits an efficient calculus. For this proxy, we establish $\mathbb{E}|\tilde F_L|^4\gg(\mathbb{E}|\tilde F_L|^2)^2$ and $\mathbb{E}({\rm det}\tilde F_L-1)^2\ll 1$. In view of the former property, we assimilate this phenomenon to intermittency. In fact, $\tilde F_L$ behaves like a tensorial stochastic exponential, and as a field can be assimilated to multiplicative Gaussian chaos.

Asymptotic equivalence of discretely observed diffusion processes and their Euler scheme: small variance case

The Euler-Maruyama Approximation of State-Dependent Regime Switching Diffusions

Variable-Step Euler–Maruyama Approximations of Regime-Switching Jump Diffusion Processes

On the convergence order of the Euler scheme for scalar SDEs with Hölder-type diffusion coefficients

Optimal finite-differences discretization for the diffusion equation from the perspective of large-deviation theory

Strong convergence of the exponential Euler scheme for SDEs with superlinear growth coefficients and one-sided Lipschitz drift

On diffusion approximation with discontinuous coefficients

A critical drift-diffusion equation: intermittent behavior

Equilibrium with coordinate dependent diffusion: Comparison of different stochastic processes

Tamed-adaptive Euler-Maruyama approximation for SDEs with superlinearly growing and piecewise continuous drift, superlinearly growing and locally Hölder continuous diffusion

The Euler scheme for stochastic differential equations with discontinuous drift coefficient: a numerical study of the convergence rate

Nonparametric estimation of the diffusion coefficient from S.D.E. paths

On the Rate of Convergence of Weak Euler Approximation for Nondegenerate Itô Diffusion and Jump Processes

Convergence of a Metropolized Integrator for Stochastic Differential Equations with Variable Diffusion Coefficient

On Asymptotic Preserving schemes for a class of Stochastic Differential Equations in averaging and diffusion approximation regimes

Convergence to the uniform distribution of moderately self-interacting diffusions on compact Riemannian manifolds

Large deviations at level 2.5 and for trajectories observables of diffusion processes : the missing parts with respect to their random-walks counterparts

Asymptotic stability of switching diffusions having sub-exponential rates of decay

On the $\varepsilon$-Euler-Maruyama scheme for time inhomogeneous jump-driven SDEs

Asymptotic equivalence of probability measures and stochastic processes

Malliavin calculus for the optimal estimation of the invariant density of discretely observed diffusions in intermediate regime