🔢 mathematics

Hydrodynamic limit from nonlinear Fokker--Planck to barotropic Euler equations

This paper establishes the hydrodynamic limit from a kinetic nonlinear Fokker-Planck equation with degenerate diffusion to the barotropic Euler equations with power-law pressure, extending previous results on isothermal systems through relative entropy analysis and generalized Log-Sobolev inequalities.

Original authors: José A. Carrillo Carrillo, Dowan Koo

Published 2026-06-30

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: José A. Carrillo Carrillo, Dowan Koo

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a massive, chaotic crowd of people (the gas particles) moving in a giant room. Each person has their own speed and direction, bumping into neighbors and reacting to the environment. This is the microscopic view: a messy, complex dance of billions of individuals.

Now, imagine you step back and look at the crowd from a helicopter. You can't see the individuals anymore; you only see the macroscopic flow: the density of the crowd in different areas and the average direction they are moving. This is the fluid view.

This paper is about mathematically proving how the chaotic, individual dance of the crowd inevitably smooths out into a predictable, flowing river when you look at it from far enough away. Specifically, it connects two different mathematical descriptions of this process.

The Two Descriptions

The "Fokker-Planck" Equation (The Microscopic View):
Think of this as a rulebook for every single person in the crowd. It describes how they move, how they bump into each other, and how they spread out. In this paper, the authors look at a special, complex version of this rulebook where the "spreading out" (diffusion) isn't simple or uniform. It's nonlinear and degenerate, meaning the way people spread out changes depending on how crowded they already are. If the crowd is thin, they spread one way; if it's thick, they spread another.
The "Barotropic Euler" Equations (The Macroscopic View):
This is the rulebook for the crowd as a whole fluid. It describes the density of the crowd and its velocity. A key part of this rulebook is the pressure: how much the crowd pushes back when you try to squeeze it. The paper focuses on a specific type of pressure called power-law pressure (where the push-back gets stronger as the crowd gets denser, following a specific mathematical curve).

The Big Question

The authors wanted to answer: If we start with the complex, individual rulebook (Fokker-Planck) and let time pass, does the crowd naturally evolve to follow the simple, fluid rulebook (Euler)?

In physics, there is a small parameter called $\epsilon$ (epsilon). You can think of this as a "zoom level" or a "time scale."

When $\epsilon$ is large, you see the messy individual movements.
As $\epsilon$ gets closer to zero, we are zooming out faster and faster.
The goal is to prove that as $\epsilon \to 0$ , the messy individual movements perfectly match the smooth fluid flow.

The Challenge: The "Degenerate" Diffusion

Previous studies had already solved this for simple, linear spreading (like ink dropping in water). But this paper tackles a much harder case: degenerate diffusion.

The Analogy:
Imagine the crowd is moving through a room filled with different types of fog.

Linear Diffusion: The fog is uniform. People spread out at a steady rate no matter where they are.
Degenerate Diffusion (This Paper): The fog is thick in some spots and thin in others. In very dense areas, the fog might be so thick that people can barely move, or they might move in a completely different way. The rules for spreading change based on the crowd density itself.

This makes the math extremely difficult because the usual tools don't work when the "friction" or "spreading" vanishes in certain areas.

How They Solved It: The "Relative Entropy" Method

To prove the connection, the authors used a technique called the Relative Entropy Method.

The Metaphor:
Imagine you have a "Perfect Fluid" (the ideal Euler equations) and a "Real Crowd" (the Fokker-Planck solution).

Entropy is a measure of disorder or "distance" from the perfect state.
Relative Entropy is a way to measure the "distance" between the Real Crowd and the Perfect Fluid at any given moment.

The authors' strategy was:

Define the Distance: Create a mathematical formula that measures how different the Real Crowd is from the Perfect Fluid.
Track the Distance: Show that as time goes on (and as $\epsilon$ gets smaller), this "distance" shrinks to zero.
The Hurdle: The tricky part was handling the "pressure" difference. In the fluid world, pressure is a simple formula. In the crowd world, pressure is a messy result of billions of collisions. The authors had to prove that the "messy pressure" of the crowd converges to the "simple pressure" of the fluid.

The Key Innovation: The "Extension Map"

To handle the complex, changing rules of the crowd (the degenerate diffusion), the authors invented a clever mathematical trick called an Extension Map.

The Analogy:
Imagine trying to compare two shapes: a flat, 2D circle and a 3D sphere. It's hard to compare them directly. But if you "lift" the flat circle into the third dimension, turning it into a sphere, you can compare them much more easily.

The authors did something similar. They took their complex, 2D crowd problem and mathematically "lifted" it into a higher dimension (adding an extra variable). In this higher dimension, the messy, changing rules of the crowd looked like a simple, uniform sphere. This allowed them to use powerful mathematical inequalities (like the Log-Sobolev inequality) to prove that the "distance" between the crowd and the fluid must shrink.

The Conclusion

The paper successfully proves that:

Even with complex, density-dependent spreading rules (degenerate diffusion), the chaotic movement of particles does settle down into the smooth, predictable flow of a fluid.
The specific type of fluid pressure that emerges is the power-law pressure (like $P = \rho^\gamma$ ), which covers many real-world scenarios, from gases to certain types of liquids.
They didn't just say "it works"; they provided a quantitative estimate. They calculated exactly how fast the crowd converges to the fluid flow as the "zoom level" ( $\epsilon$ ) changes.

In short: The authors built a mathematical bridge showing that even when the microscopic rules of a gas are complicated and change based on density, the macroscopic result is always a clean, predictable fluid flow described by the Euler equations. They did this by inventing a new way to "lift" the problem into a higher dimension to make the math tractable.

Technical Summary: Hydrodynamic Limit from Nonlinear Fokker–Planck to Barotropic Euler Equations

Problem Statement
This work addresses the rigorous derivation of the hydrodynamic limit for a kinetic nonlinear Fokker–Planck equation with degenerate diffusion. Specifically, the authors consider the equation:
$\partial_t f + v \cdot \nabla_x f = \frac{1}{\varepsilon} \nabla_v \cdot (\nabla_v L_\psi(f) + (v - u_f)f)$
where $L_\psi$ represents a nonlinear diffusion law and $\varepsilon$ is a scaling parameter tending to zero. The objective is to prove that as $\varepsilon \to 0$ , the macroscopic quantities (density $\rho$ and momentum $m = \rho u$ ) derived from the kinetic distribution $f$ converge to the solution of the compressible barotropic Euler equations:
$\begin{cases} \partial_t \rho + \nabla_x \cdot (\rho u) = 0, \\ \partial_t (\rho u) + \nabla_x \cdot (\rho u \otimes u) + \nabla_x P(\rho) = 0, \end{cases}$
with a general pressure law $P(\rho)$ , including the power-law case $P(\rho) = \theta \rho^\gamma$ .

While the derivation of isothermal Euler equations ( $P(\rho) \propto \rho$ ) from Fokker–Planck equations with linear diffusion is well-established (e.g., Berthelin and Vasseur [3]), and isentropic limits via BGK models have been studied (e.g., Bouchut [4]), the derivation of general barotropic Euler equations from Fokker–Planck equations with degenerate nonlinear diffusion had not been previously addressed.

Methodology
The analysis relies on the relative entropy method, a standard technique for establishing weak-strong uniqueness and convergence rates in hydrodynamic limits. The core technical strategy involves:

Kinetic-Macroscopic Correspondence: The authors establish a precise link between the microscopic nonlinear diffusion law $L_\psi$ and the macroscopic pressure law $P_\psi$ . They define a macroscopic enthalpy $h_\psi$ via an integral equation involving the inverse of the diffusion function $\psi^{-1}$ . This allows them to characterize the equilibrium state (local Maxwellian) $M_\psi[f]$ and the associated pressure tensor.
Generalized Log-Sobolev Inequality: To control the dissipation of the kinetic model, the authors employ a generalized Log-Sobolev inequality (citing [8]) tailored for degenerate diffusion. This inequality bounds the relative entropy (the difference between the kinetic free energy and the equilibrium free energy) by the kinetic dissipation term.
Extension Map and Radialization: A critical technical contribution is the control of the pressure error tensor. The authors adapt the "extension map" technique (inspired by [17]) and rearrangement ideas (from [3]). They extend the distribution function $f$ from $\mathbb{R}^N$ to $\mathbb{R}^{N+1}$ using a characteristic function on a half-space. By projecting this extended function onto a one-dimensional radial variable, they decompose the error into "local" (near the equilibrium radius) and "global" parts.
Dissipation Control: Using the extended framework, they prove a quantitative estimate (Proposition 5.1) showing that the $L^1$ error between the kinetic pressure tensor and the macroscopic pressure $P_\psi(\rho)$ is bounded by the square root of the relative entropy and the relative entropy itself. This step requires a doubling condition on the diffusion law and a specific auxiliary lemma (Lemma 5.4) concerning integrals of functions satisfying doubling conditions.

Key Contributions

Generalization of Pressure Laws: The paper extends the hydrodynamic limit results from isothermal ( $\gamma=1$ ) to general barotropic equations, including isentropic cases ( $\gamma > 1$ ) and pressureless gas dynamics ( $\gamma=0$ ), derived from a Fokker–Planck equation with degenerate nonlinear diffusion.
Admissible Diffusion Laws: The authors define a broad class of admissible nonlinear diffusion laws $L_\psi$ (satisfying assumptions A1–A5) that correspond to a wide range of pressure laws, including power laws $P(\rho) = \theta \rho^\gamma$ and pressures with different asymptotic behaviors at low and high densities.
Technical Innovation in Error Estimates: The paper introduces a carefully designed extension map and radialization procedure to handle general nonlinear diffusion. This allows for the control of pressure tensor errors without restricting the diffusion to specific power-law forms, overcoming limitations of previous works that relied on specific BGK models or linear diffusion.
Quantitative Convergence Rates: Under the assumption of well-prepared initial data and the existence of a classical solution to the Euler equations, the authors derive explicit convergence rates for the macroscopic quantities in terms of $\varepsilon$ .

Main Results
The primary result is Theorem 6.1, which states that if the initial data is well-prepared (specifically, the initial relative entropy is of order $\varepsilon^{1/2}$ ), then the relative entropy between the kinetic solution and the macroscopic Euler solution remains bounded by $C\varepsilon^{1/2}$ uniformly in time.

Consequently, the macroscopic quantities converge strongly:

$\rho_\varepsilon \to \rho$ and $\rho_\varepsilon u_\varepsilon \to \rho u$ in $L^\infty([0, T]; L^1_{loc})$ .
The pressure tensor converges: $\int (v-u_\varepsilon)\otimes(v-u_\varepsilon)f_\varepsilon dv \to P_\psi(\rho)I_N$ in $L^1_{loc}$ .

For the specific case of power-law pressure $P(\rho) = \theta \rho^\gamma$ , Corollary 6.3 provides explicit convergence rates:

Density error: $\|\rho_\varepsilon - \rho\|_{L^\infty L^\gamma} \leq C \varepsilon^\beta$
Momentum error: $\|\rho_\varepsilon u_\varepsilon - \rho u\|_{L^\infty L^{2\gamma/(\gamma+1)}} \leq C \varepsilon^\beta$
Flux and pressure tensor errors follow similar rates, where $\beta = \frac{1}{2 \max\{2, \gamma\}}$ .

Significance and Claims
The authors position this work as filling a gap in the literature regarding the derivation of isentropic and general barotropic Euler equations from Fokker–Planck type equations with degenerate diffusion. While previous studies focused on linear diffusion (isothermal) or specific BGK models, this work demonstrates that a broad class of barotropic Euler equations arises naturally as the hydrodynamic limit of nonlinear Fokker–Planck equations.

The paper explicitly notes that the existence of solutions to the kinetic equation (1.1) is not treated within this work; thus, the hydrodynamic limit result is "a priori." However, the authors reference recent work [6] that addresses the gradient flow structure and solution construction for a specific case ( $u_f=0$ ), suggesting the framework is consistent with existing existence theories. The significance lies in the rigorous mathematical justification of the macroscopic pressure law emerging from microscopic nonlinear diffusion mechanisms, utilizing a robust relative entropy framework enhanced by novel geometric extension techniques.