Quantitative estimates for flows of regular Lagrangian flows for Hörmander singular kernels and LD vector fields

Basic Information

• Paper ID: 2510.13690
• Title: Quantitative estimates for flows of regular Lagrangian flows for Hörmander singular kernels and LD vector fields
• Author: Henrique Borrin (University of São Paulo, Ribeirão Preto Campus, Brazil)
• Classification: math.AP (Analysis of PDEs), math.DS (Dynamical Systems)
• Submission Date: October 15, 2025
• Paper Link: https://arxiv.org/abs/2510.13690v1

Abstract

This paper investigates quantitative estimates for regular Lagrangian flows associated with vector fields whose derivatives can be expressed as convolutions of fundamental singular kernels satisfying "Hörmander" conditions with space-time integrable functions.

Research Background and Motivation

Problem Background

The core problem studied concerns the well-posedness of flow equations: $\frac{d}{dt}X(t,s,x) = b(t,X(t,s,x)), \quad X(s,s,x) = x$

where the vector field $b:[0,T] \times \mathbb{R}^d \to \mathbb{R}^d$ satisfies growth assumptions: $\frac{b(t,x)}{1+|x|} = b_1(t,x) + b_2(t,x)$ with $b_1 \in L^1([0,T] \times \mathbb{R}^d)$ and $b_2 \in L^1([0,T]; L^{\infty}(\mathbb{R}^d))$.

Research Motivation

1. Theoretical Requirements: Well-posedness of flow equations for non-smooth vector fields is a fundamental problem in PDE theory with important applications to transport equations and continuity equations.
2. Technical Challenges: Traditional approaches are limited to $L^1((0,T); W^{1,1}_{loc}(\mathbb{R}^d))$ vector fields, while this paper addresses the more general class of $L^1((0,T); LD_{loc}(\mathbb{R}^d))$ vector fields.
3. Methodological Innovation: Existing literature has incomplete treatment of Lagrangian variations, particularly regarding singular kernel handling.

Core Contributions

1. Generalization of Fundamental Estimates: Extends the Bouchut-Crippa result to more general classes of Hörmander singular kernels, no longer restricted to $C^1$ kernels.
2. Extension of Vector Field Classes: Extends from $W^{1,1}_{loc}$ vector fields to $LD_{loc}$ vector fields, where $LD(\Omega) = \{u \in L^1(\Omega;\mathbb{R}^d) : \partial_j u^i + \partial_i u^j \in L^1(\Omega)\}$
3. Establishment of Quantitative Estimates: Proves fundamental estimates for regular Lagrangian flows, ensuring well-posedness.
4. Provision of Applications: Proves existence of solutions to transport equations and continuity equations.

Methodology Details

Problem Formulation

Studies vector fields $b$ satisfying symmetric derivative conditions: $\partial_j b^i(t,x) + \partial_i b^j(t,x) = \sum_{k=1}^m \Gamma^{ijk,R} * g^{ijk,R}(t,x)$ where $\Gamma^{ijk,R}$ are fundamental singular kernels and $g^{ijk,R} \in L^1([0,T] \times \mathbb{R}^d)$.

Core Definitions

Definition 1.1 (Fundamental Singular Kernel): A function $\Gamma \in L^1_{loc}(\mathbb{R}^d \setminus \{0\})$ with $\mathcal{F}(\Gamma) \in L^{\infty}(\mathbb{R}^d)$ is called a fundamental singular kernel if: $C_0 := \sup_{z \neq 0} \int_{|y|>2|z|} |\Gamma(y-z) - \Gamma(y)| dy < \infty$

Definition 1.2 (Regular Lagrangian Flow): A map $X:[s,T] \times [0,T] \times \mathbb{R}^d \to \mathbb{R}^d$ is called a regular Lagrangian flow if:

1. It satisfies regularity conditions $X \in C(D_T; L^0_{loc}(\mathbb{R}^d)) \cap B(D_T; \log(L)_{loc}(\mathbb{R}^d))$
2. It satisfies the renormalized flow equation
3. There exists a compression constant $L > 0$

Main Results

Theorem 1.1 (Fundamental Estimate): Let $b, \bar{b}$ be vector fields satisfying the stated conditions, and $X, \bar{X}$ be corresponding regular Lagrangian flows. Then for any $\gamma, r, \eta > 0$, there exist $\lambda > 0$ and a constant $C_{\gamma,r,\eta} > 0$ such that: $\mathcal{L}^d(B_r \cap \{x \in \mathbb{R}^d : |X(t,s,x) - \bar{X}(t,s,x)| > \gamma\}) < C_{\gamma,r,\eta}\|b - \bar{b}\|_{L^1((s,\tau) \times B_{\lambda})} + \eta$

Technical Innovations

1. Singular Kernel Theory: Employs Hörmander conditions instead of pointwise decay conditions, significantly extending applicability.
2. Maximal Operators: Introduces maximal operators $M_{\{\rho_\nu\}}$ for handling convolution estimates: $M_{\{\rho_\nu\}} u(x) := \sup_\nu \sup_{\epsilon>0} |\rho_\nu^\epsilon * u(x)|$
3. Symmetric Derivative Techniques: Utilizes the symmetric part of derivatives $\text{Sym}(Du)$ rather than full derivatives, reducing regularity requirements.

Experimental Setup

Theoretical Verification

This is pure theoretical work, verified through:

1. Comparison with Existing Results: Demonstrates that the paper's results completely generalize Bouchut-Crippa's results
2. Special Case Verification: Verifies recovery of known results under classical assumptions
3. Application Verification: Validates theory through well-posedness of transport and continuity equations

Technical Tools

1. Calderón-Zygmund Decomposition: For handling singular integral operators
2. Sublevel Set Control: For controlling flow growth properties
3. Renormalization Techniques: For treating non-smooth vector fields

Main Technical Results

Weak $L^1$ Estimates

Proposition 3.1: For singular kernel $\Gamma$ and function $u \in L^1(\mathbb{R}^d)$: $|||M_{\{\rho_\nu\}} \Gamma * u|||_{L^1_w(\mathbb{R}^d)} \leq C\|u\|_{L^1(\mathbb{R}^d)}$

Mean Value Inequalities

Proposition 4.1: For $LD$ vector field $u$, there exist various maximal operators such that: $(u(x) - u(y)) \cdot (x-y) \leq |x-y|^2 \sum_{\text{operators}} [M_{\text{op}}(\text{Sym}(Du))(x) + M_{\text{op}}(\text{Sym}(Du))(y)]$

Related Work

Main Development Timeline

1. DiPerna-Lions Theory: Pioneering work on flow theory for $W^{1,1}$ vector fields
2. Ambrosio et al. BV Theory: Extension to BV vector fields
3. Bouchut-Crippa: Introduction of singular integral methods
4. Nguyen's Recent Work: Treatment of BV case with stronger structural assumptions

Position of This Work

This paper neither contains nor is contained in Nguyen's work; each has distinct advantages:

• This Paper's Advantage: Handles more general classes of singular kernels
• Nguyen's Advantage: Handles finite measure settings

Conclusions and Discussion

Main Conclusions

1. Establishes regular Lagrangian flow theory for Hörmander singular kernels and LD vector fields
2. Proves fundamental quantitative estimates ensuring flow well-posedness
3. Applies to transport and continuity equations, proving solution existence

Limitations

1. Cannot handle finite measure settings (gap with Nguyen's work)
2. Still requires vector fields to satisfy certain growth conditions
3. Boundedness requirement on Fourier transform of singular kernels may be overly restrictive

Future Directions

1. Extension to finite measure settings
2. Investigation of more general growth conditions
3. Application to specific physical models (e.g., Vlasov equations)

In-Depth Evaluation

Strengths

1. Theoretical Innovation: Successfully generalizes classical results to more general settings
2. Technical Depth: Cleverly combines singular integral theory with flow theory
3. Clear Exposition: Well-structured with detailed proofs
4. Applied Value: Provides new tools for transport equation theory

Weaknesses

1. Scope of Applicability: Some technical assumptions may be overly restrictive
2. Computational Complexity: Involves multiple maximal operators, potentially difficult for practical application
3. Relation to Related Work: Comparison with Nguyen's work could be more thorough

Impact

1. Theoretical Contribution: Makes important contributions to non-smooth vector field theory
2. Methodology: Provides new techniques for handling singular kernels
3. Application Prospects: Potential applications in fluid mechanics and dynamical systems

Applicable Scenarios

1. Analysis of vector fields with singular kernel structure
2. Theoretical research on transport and continuity equations
3. Lagrangian descriptions in fluid mechanics

References

The paper cites important literature in the field, including:

• Pioneering work by DiPerna-Lions
• Contributions by Ambrosio on BV vector fields
• Singular integral methods by Bouchut-Crippa
• Recent advances by Nguyen

Technical Assessment: This is a high-quality theoretical mathematics paper making important contributions at the intersection of partial differential equations and dynamical systems. The author successfully generalizes classical flow theory to more general settings with refined technical treatment and results of significant theoretical value.