The doubling property of the elliptic measure, for elliptic operators with drifts satisfying an average diverging condition
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.
Ada: We're looking at the paper: "The doubling property of the elliptic measure, for elliptic operators with drifts satisfying an average diverging condition".
Emmy: Here's the gist of the paper. The paper shows that doubling of the elliptic measure corresponding to an operator with an elliptic principal term and a drift that diverges, on average on Whitney cubes, like the inverse distance to the boundary, with a small constant.
Ada: Pruning redundant reasoning tokens We show doubling of the elliptic measure corresponding to the operator with an elliptic principal term and a drift that diverges, on average on Whitney cubes, like the inverse distance to the boundary, with a small constant.
Emmy: This generalizes earlier results with the hypothesis of pointwise smallness of such a drift.
Ada: How the training signal is built The paper considers a 1-sided chord arc domain with an operator Lu = −∇ ⋅ (A∇u) + B ⋅ ∇u.
Title and authors: Ada: We just talked about the paper "The doubling property of the elliptic measure, for elliptic operators with drifts satisfying an average diverging condition." It focuses on how an operator with a drifting term still maintains a doubling property for its measure when that drift satisfies an average diverging condition.
Emmy: Essentially, they are showing that this average divergence condition is enough to establish this doubling, which we know from earlier work under pointwise smallness assumptions >
Kolya: So the main idea here seems to be broadening the applicability of existing solvability results by relaxing a strong assumption on the drift >
Ada: That’s right. They are using this average divergence condition instead of pointwise smallness, and they link it to results from Hofmann and Lewis and Kenig Pipher concerning rough Dirichlet solvability >
Emmy: And they also use this result to prove Hardy inequalities in specific geometric settings, like one-sided chord arc domains, where the boundary has an Ahlfors David regularity > <ref:2511.12942#pg1,1-sided chord arc domains>
Kolya: It seems like a lot of machinery is going into proving that the Green's function bounds are robust enough for these rough problems. What about the limitations they mention?
Ada: The paper notes that in their setting, they assume solutions to L and its adjoint L star exist before proving the Green's function estimates >
Emmy: That’s a practical limitation for applying it directly; we need those solutions to be guaranteed first, which isn't always the case in rough problems >
The paper's summary: Ada: To summarize, this paper tackles the doubling property of elliptic measures for operators with drifts that satisfy an average diverging condition. They show that this condition is sufficient to get doubling on average over Whitney cubes, related to the inverse distance to the boundary, with a small constant >
Emmy: The key mechanism they use is combining a small Carleson constant assumption on the drift and a background pointwise bound with a large constant >
Kolya: So when you put those two things together, what’s the tangible payoff for someone working on these rough Dirichlet problems?
Ada: The payoff is that they get two-sided bounds on the Green's function, which are vital for getting pointwise upper and lower estimates for that function, which enables them to tackle rough Dirichlet problems >
Emmy: And this leads directly to boundary Hölder continuity for the solution u under certain conditions on the drift term >
Kolya: So it’s a chain: average divergence condition leads to measure doubling, which leads to Green's function bounds, which then lets us control the solution better near boundaries >
Ada: That’s a good way to put it. It moves us from a problem where we needed pointwise smallness for the drift to one where we only need an average divergence condition >
The paper's improvements: Ada: The authors suggest some improvements, and they focus on making the connection more robust. They want to enhance the solvability analysis by combining averaged smallness conditions with pointwise bounds >
Emmy: They argue that this combination allows AI systems to reliably establish Bourgain type estimates and prove L p Dirichlet solvability even when the drift doesn't satisfy a pointwise smallness assumption >
Kolya: That means they are trying to make their method more universally applicable, moving beyond just specific cases where the drift is pointwise small >
Ada: Exactly. And they also want improved control over Green's function behavior in non-divergence form operators by leveraging the doubling of that elliptic measure as a tool >
Emmy: This doubling allows for the development of robust pointwise upper and lower bounds for G(y, x) in domains with drifts, which is crucial when solving rough Dirichlet problems >
Kolya: So these improvements are about making the tools more versatile across different types of operators and drift conditions >
Conclusion: Ada: To wrap up, this paper "The doubling property of the elliptic measure, for elliptic operators with drifts satisfying an average diverging condition" shows that the average divergence condition is sufficient to establish measure doubling under those conditions >
Emmy: This result gives us robust two-sided bounds on the Green's function and boundary Hölder continuity for solutions under certain drift conditions >
Kolya: And these improvements point towards making methods more versatile by combining averaged smallness with pointwise bounds to get better solvability estimates >
Ada: Ultimately, this work provides a framework where we don't need to assume the drift is pointwise small for the doubling property, which is a big step in understanding rough Dirichlet problems >
Emmy: It’s about showing that average divergence conditions are enough to control the measure structure without needing that strong pointwise constraint on every single point >
Kolya: So we're looking at this paper as a way to make the tools we have more applicable across a wider range of operators and drift conditions, right?
Ada: Exactly. The paper "The doubling property of the elliptic measure, for elliptic operators with drifts satisfying an average diverging condition" is a significant piece for anyone studying rough Dirichlet problems >
Aritro Pathak
math.AP
Submitted: 2025-11-17
Updated: 2026-11-09
Comments: Corrected a minor error with addition of Lemma 6
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 83/100
The gist: The paper shows that doubling of the elliptic measure corresponding to an operator with an elliptic principal term and a drift that diverges, on average on Whitney cubes…
Key concepts
- Elliptic Measure Doubling
- This property means that as you scale regions within an elliptic operator's domain, the measure associated with it grows predictably. Specifically, doubling the size of a region results in a corresponding increase in the measure by a small constant factor. This is crucial for controlling solutions and establishing regularity properties.
- Drift Term Divergence
- The paper deals with operators that have a drift term (like B ⋅ ∇u) that does not vanish pointwise but diverges on average over specific regions, such as Whitney cubes. This condition is more general than requiring the drift to be small everywhere, allowing the analysis to handle rougher coefficients.
- Whitney Cube Decomposition
- This is a way to partition a domain into smaller, well-behaved geometric pieces called Whitney cubes. These cubes are used in the analysis because they provide local control over the geometry and allow for localized estimates of the operator's behavior, which is essential for proving global properties.
- Green's Function Bounds
- The Green's function describes how solutions behave near boundaries or singularities. The paper derives two-sided bounds on this function, especially in relation to the elliptic measure doubling. These bounds are vital for proving pointwise estimates and ensuring that solutions have controlled behavior, even when the operator has a drift.
Terminology
Summary
The paper shows that doubling of the elliptic measure corresponding to an operator with an elliptic principal term and a drift that diverges, on average on Whitney cubes, like the inverse distance to the boundary, with a small constant. p. 0 The doubling property of the elliptic measure, for elliptic
operators with drifts satisfying an average diverging condition. Aritro Pathak
July 2, 2026
Abstract
We show doubling of the elliptic measure corresponding to the operator with an elliptic principal term and a drift that diverges, on average on Whitney cubes, like the inverse distance to
the boundary, with a small constant.Where the paper says it
Pruning redundant reasoning tokens
We show doubling of the elliptic measure corresponding to the operator with an elliptic principal term and a drift that diverges, on average on Whitney cubes, like the inverse distance to the boundary, with a small constant. This generalizes earlier results with the hypothesis of pointwise smallness of such a drift. p. 0 Aritro Pathak
July 2, 2026
Abstract
We show doubling of the elliptic measure corresponding to the operator with an elliptic principal term and a drift that diverges, on average on Whitney cubes, like the inverse distance to
the boundary, with a small constant. Essentially a small Carleson constant assumption on the
drift, this generalizes earlier results with the hypothesis of pointwise smallness of such a drift.Where the paper says it
How the training signal is built
The paper considers a 1-sided chord arc domain with an operator Lu = −∇ ⋅ (A∇u) + B ⋅ ∇u. The drift term is assumed to be small in an averaged sense over each Whitney region, meaning there exists a β small enough so that for each Whitney cube UQ in the Whitney cube decomposition of omega we have (1) ∶ ∫ UQ ∣B∣ 2 δ(X)dV ≤ β2 l(Q) n−1. p. 0 Consider a 1-sided chord arc domain Ω ⊂ R
n with n ≥ 3. Consider the linear operator
Lu = −∇ ⋅ (A∇u) + B ⋅ ∇u, (1)
We assume the drift term is small in an averaged sense over each Whitney region, which is to say that
there exists a β small enough so that for each Whitney cube UQ in the Whitney cube decomposition
of Ω we have,
(1) ∶ ∫
UQ
∣B∣
2
δ(X)dV ≤ β
2
l(Q)
n−1
, (2)
Here ∣B∣ denotes the magnitude of the drift term. Consider the linear operator
Lu = −∇ ⋅ (A∇u) + B ⋅ ∇u, (1)
We assume the drift term is small in an averaged sense over each Whitney region, which is to say that
there exists a β small enough so that for each Whitney cube UQ in the Whitney cube decomposition
of Ω we have,
(1) ∶ ∫
UQ
∣B∣
2
δ(X)dV ≤ β
2
l(Q)
n−1
, (2)
Here ∣B∣ denotes the magnitude of the drift term. In this paper, this is a generalization of the case
considered in Chapter I of [HL01] where we had the pointwise smallness estimate of the form
∣B(x)∣ ≤
β
δ(x)
, (3)
for some small enough β.Where the paper says it
Where the paper says it
Where the gains come from
The doubling property is established by combining two ingredients, which are of independent interest. These ingredients are a small Carleson constant assumption on the drift and background pointwise bound with the large constant as in Eq. p. 2 This paper establishes doubling of the elliptic measure for this more general situation of drifts that
have a smallness condition Eq. (2) on average on every whitney cube in a chord arc domain, instead
of having a pointwise smallness assumption, along with a background pointwise bound with the large
constant as in Eq. Essentially a small Carleson constant assumption on the
drift, this generalizes earlier results with the hypothesis of pointwise smallness of such a drift. This relates to recent perturbative results of rough Dirichlet solvability in domains with drifts or
potentials that satisfy a Carleson measure condition, which have also been considered earlier by
Hofmann-Lewis and Kenig-Pipher. The
Carleson measure condition of Eq. (7) implies with a standard pigeonholing argument that, for a given
small ϵ > 0, for all but a finite number of Whitney cubes contained in B(x, r) ,
∫
UQ
sup
B(t,δ(t)/2)
(∣B∣
2
(y)δ(y))dt < ϵl(Q)
n−1
.Where the paper says it
Where the paper says it
Where the paper says it
Theorem 3 establishes a local boundary Hardy inequality for a function u in Y12 (omega)∩C(omega), with u∣∂omega = 0, and the boundary ∂omega is (n − 1)-dimensional Ahlfors David regular. The proof uses the Whitney decomposition of omega, the Poincare inequality, along with a dyadic version of Hardy’s inequality. p. 5 Proof of Theorem 3. The proof follows by the use of the Whitney decomposition of Ω, the Poincare
inequality, along with a dyadic version of Hardy’s inequality along with the continuity of u at the
boundary.Where the paper says it
Theorem 5 establishes boundary H¨older continuity for the solution u to Eq. (1) under certain conditions on the drift term. p. 1 In a
separate manuscript, we are studying the operator Eq. (1) with the conditions of Eqs.Where the paper says it
The final result, Corollary 6 (Bourgain estimate), is obtained by obtaining two-sided bounds on the Green function from Theorems 9 and 10, which immediately give us the doubling for the elliptic measure.
Theorem 8 provides a pointwise lower bound for the Green function G(y, z) when y − z ≤ 1/2 δ(z). This is a consequence of Harnack inequality and Theorem 8. p. 2 In
[Hof25] we prove the local version of Theorem 3 as a consequence of this global version. However, a
direct proof of this local (and thus also the global) Hardy inequality using a stopping time argument
here is interesting in its own right, and to the best knowledge of the author hasn’t directly appeared
3 In
[Hof25] we prove the local version of Theorem 3 as a consequence of this global version. However, a
direct proof of this local (and thus also the global) Hardy inequality using a stopping time argument
here is interesting in its own right, and to the best knowledge of the author hasn’t directly appeared
3Where the paper says it
Where the paper says it
Theorem 7 establishes the Boundary Harnack inequality for the elliptic operator in Eq. (1) in the setting of a 1-sided chord arc domain. p. 2 (2) on the drift, the arguments become more delicate, in the general setting of the 1−sided chord
arc domains. In [HMT17], it is pointed out that Dirchlet solvability holds by weakening the hypothesis of Eq. Proof of Theorem 3. The proof follows by the use of the Whitney decomposition of Ω, the Poincare
inequality, along with a dyadic version of Hardy’s inequality along with the continuity of u at the
boundary.Where the paper says it
Where the paper says it
Theorem 8 provides the pointwise upper bounds in such domains bounded away from the boundary. This relies on a Calderon Zygmund decomposition and careful control of terms involving the drift term. p. 1 The pointwise estimates on the Green function
with lower order terms is itself of interest in other areas of partial differential equations[Mour23,
Sak21, KS19]. Although the lower pointwise estimates are proved by adopting the techniques of
[GW82], the upper pointwise estimates require more work, and crucially needs a Calderon Zygmund
type decomposition of the Whitney cubes.Where the paper says it
The proof of Theorem 8 involves using the Green function for uniformly elliptic equations. This leads to the bound inf y∈By,x∈Bx G(x, y) ≤ C x − y (n−2). p. 9 We can write the second term on the right as,
∫
UQ
∣B∣
2
u
2ψ
2
dV ≤ (sup
UQ
u)
2
(sup
UQ
ψ)
2
∫
UQ
∣B∣
2
dV. (51)
We claim that there exist uniform constants η1 ≥ 1, 0 < η2 < 1 so that, for the Whitney cube UQ
we have on an ample subset U
′
Q ⊂ (supp(ψ) ∩ UQ) with the property ∣U
′
Q∣/∣supp(ψ) ∩ UQ∣ ≥ η2
and thus ∣U
′
Q∣/∣UQ∣ ≥ η2θ, that (sup
U′
Q
ψ)/(inf
U′
Q
ψ) < η1. We can write the second term on the right as,
∫
UQ
∣B∣
2
u
2ψ
2
dV ≤ (sup
UQ
u)
2
(sup
UQ
ψ)
2
∫
UQ
∣B∣
2
dV. (51)
We claim that there exist uniform constants η1 ≥ 1, 0 < η2 < 1 so that, for the Whitney cube UQ
we have on an ample subset U
′
Q ⊂ (supp(ψ) ∩ UQ) with the property ∣U
′
Q∣/∣supp(ψ) ∩ UQ∣ ≥ η2
and thus ∣U
′
Q∣/∣UQ∣ ≥ η2θ, that (sup
U′
Q
ψ)/(inf
U′
Q
ψ) < η1.Where the paper says it
Where the paper says it
Theorem 9 establishes the lower bound for the Green function G(y, z) for the operator in Eq. (1) when coefficients satisfy Eq. p. 0 Essentially a small Carleson constant assumption on the
drift, this generalizes earlier results with the hypothesis of pointwise smallness of such a drift. This relates to recent perturbative results of rough Dirichlet solvability in domains with drifts or
potentials that satisfy a Carleson measure condition, which have also been considered earlier by
Hofmann-Lewis and Kenig-Pipher. This global version is well known from earlier results going back at least to [Lew88]. In
[Hof25] we prove the local version of Theorem 3 as a consequence of this global version.Where the paper says it
Where the paper says it
Theorem 6 shows that for any x ∈ omega, and xˆ a nearest point to x on the boundary, with x−xˆ = r, there exists a uniform constant c > 0, we have, with ∆(x) ∶= ∆(x, 10r), that ω x (∆(x)) ≥ c. This is obtained by using the boundary H¨older continuity for the function v = 1 − u which vanishes continuously on the surface ball ∆(x). p. 4 (1) with
the corresponding elliptic measure ω and Green’s function G(⋅, z) with pole at z; for any x ∈ ∂Ω, r ≤
diam ∂Ω, with A(x, r) a corkscrew point relative to the surface ball ∆(x, r) ∶= B(x, r) ∩ ∂Ω, then for
any y ∈ Ω ∖ B(x, 2r), we have
c
−1
r
n−2G(y, A(x, r)) ≤ ω(y, ∆(x, r)) ≤ ω(y, ∆(x, 2r)) ≤ c
2
r
n−2G(y, A(x, r)). (19)
The Boundary H¨older regularity and subsequently the Bourgain property is stated and proved in
Section 4, and the pointwise upper and lower estimates on the Green function are proved in Section
5.Where the paper says it
Theorem 5 provides a Bourgain-type estimate in this case. This is derived from the doubling argument of [Ai06]. p. 1 (3) assumption, is also found in Chapter 1
of [HL01]. These two ingredients combine to establish the doubling of the elliptic measure, by an argument
found in [Ai06], as mentioned in Theorem 2 of this paper.Where the paper says it
Theorem 7 establishes the Boundary Harnack inequality in the setting of a 1-sided chord arc domain follows from Theorem 5 as well as the Harnack inequality. This is a consequence of Lemma 4. p. 2 This global version is well known from earlier results going back at least to [Lew88]. In
[Hof25] we prove the local version of Theorem 3 as a consequence of this global version. This global version is well known from earlier results going back at least to [Lew88]. In
[Hof25] we prove the local version of Theorem 3 as a consequence of this global version. This global version is well known from earlier results going back at least to [Lew88]. In
[Hof25] we prove the local version of Theorem 3 as a consequence of this global version. This global version is well known from earlier results going back at least to [Lew88]. In
[Hof25] we prove the local version of Theorem 3 as a consequence of this global version.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Theorem 8 provides the lower bound for the Green function for the operator in Eq. (1) for any z, y ∈ omega with y − z ≤ 1/2 δ(z). This is a consequence of Harnack inequality and Theorem 8.
Theorem 7 establishes the pointwise upper and lower bounds on the Dirichlet Green’s function in the ball B(x, δ(x)/2). This is obtained by using Eq. p. 0 While we work in 1-sided chord arc domains, these results are
new even for the half-space. In the process, we also prove Hardy inequalities in such domains
with Ahlfors-David regular boundary, using a stopping time argument.Where the paper says it
Theorem 8 proves the pointwise upper bounds in such domains bounded away from the boundary. This involves a Calderon Zygmund decomposition and careful control of terms involving the drift term. p. 1 The pointwise estimates on the Green function
with lower order terms is itself of interest in other areas of partial differential equations[Mour23,
Sak21, KS19]. Although the lower pointwise estimates are proved by adopting the techniques of
[GW82], the upper pointwise estimates require more work, and crucially needs a Calderon Zygmund
type decomposition of the Whitney cubes.Where the paper says it
The doubling for the elliptic measure is obtained by Theorem 2, which follows immediately from the argument for more general John domains, given in Lemma 3.5 and Lemma 3. p. 8 We first show boundary H¨older continuity holds with our singular drift term. This is an adaptation
of the proof of Lemma 3. (3) assumption, is also found in Chapter 1
of [HL01]. These two ingredients combine to establish the doubling of the elliptic measure, by an argument
found in [Ai06], as mentioned in Theorem 2 of this paper. In
[Hof25] we prove the local version of Theorem 3 as a consequence of this global version. However, a
direct proof of this local (and thus also the global) Hardy inequality using a stopping time argument
here is interesting in its own right, and to the best knowledge of the author hasn’t directly appeared
3Where the paper says it
Where the paper says it
Where the paper says it
The proof of Theorem 7 involves using Eq. (83) and Harnack's inequality. This is a consequence of Lemma 4.4 of [JK82].
The proof of Theorem 8 essentially follows by extending the argument of the proof of Eq.(1.9) of [GW82]. p. 1 The pointwise estimates on the Green function
with lower order terms is itself of interest in other areas of partial differential equations[Mour23,
Sak21, KS19]. Although the lower pointwise estimates are proved by adopting the techniques of
[GW82], the upper pointwise estimates require more work, and crucially needs a Calderon Zygmund
type decomposition of the Whitney cubes.Where the paper says it
The proof of Theorem 7 uses Eq. (83) and Harnack's inequality. This is a consequence of Lemma 4.4 of [JK82].
The proof of Theorem 8 uses Eq. (83) and Harnack's inequality. This is a consequence of Lemma 4.4 of [JK82].
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Enhanced solvability analysis for elliptic operators with drifts by combining averaged smallness conditions (Eq 2) with pointwise bounds (Eq 4). This allows AI systems to reliably establish Bourgain type estimates and prove L p Dirichlet solvability even when the drift does not satisfy a pointwise smallness assumption.
-
Improved control over Green's function behavior in non-divergence form operators by leveraging doubling of the elliptic measure (Theorem 2). This enables the development of robust pointwise upper and lower bounds for G(y, x) in domains with drifts, which is crucial for solving rough Dirichlet problems.
-
Development of a stopping time argument for local Hardy inequalities (Theorem 3). This capability allows AI to establish precise relationships between the integral of the square of the solution multiplied by distance to the boundary and its gradient, which is vital for proving regularity properties near boundaries.
-
Advanced boundary regularity analysis via Boundary Hölder continuity (Theorem 5). The system can now predict local bounds on solutions near a surface ball based on local information, leading to improved control over Dirichlet data and boundary behavior of solutions.
-
Refined estimates for the Dirichlet Green's function in domains with singular drifts (Theorem 10). This allows the AI to compute specific bounds for G(y, x) when y is near the boundary, enabling better approximation of solutions using Green's functions in regions bounded away from the boundary.
Abstract
We show doubling of the elliptic measure corresponding to the operator with an elliptic principal term and a drift that diverges, on average on Whitney cubes, like the inverse distance to the boundary, with a small constant. Essentially a small Carleson constant assumption on the drift, this generalizes earlier results with the hypothesis of pointwise smallness of such a drift. This relates to recent perturbative results of rough Dirichlet solvability in domains with drifts or potentials that satisfy a Carleson measure condition, which have also been considered earlier by Hofmann-Lewis and Kenig-Pipher. While we work in 1-sided chord arc domains, these results are new even for the half-space. In the process, we also prove Hardy inequalities in such domains with Alhfors-David regular boundary, using a stopping time argument.
Sources
- The Dirichlet Problem for elliptic equations with singular drift terms
- A counterexample for pointwise upper bounds on Green's function with a singular drift at boundary
Related papers
- Regularity of critical immersions for scale-invariant curvature energies in even dimensions
- C 1, alpha Regularity of W 2,2 Q-Valued Maps with Small Laplacian
- Cumulative angular cancellation in Maz'ya's endpoint inequalities
- Strictly positive optimal Hardy weights on manifolds and graphs via Green potential
- A fast dynamo with zero topological entropy
- A counterexample to fixed-profile soliton resolution for the three-dimensional quintic wave equation