Scattering of solutions to NLW by Inward Energy Decay

Ruipeng Shen

REVIEW 2 major objections 3 minor 12 references

Scattering of solutions to NLW by Inward Energy Decay

T0 review · 2 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper proves that radial, finite-energy solutions of the defocusing nonlinear wave equation in R^3 scatter in positive time whenever the inward energy decays with weight max{1,|x|^kappa} for kappa at least (5-p)/(p+1), with an…

desk verdict New scattering criterion for radial 3D NLW using only inward energy decay; proof is clean but leans at one load-bearing spot on an unproved positive-time statement from the author's earlier preprint. read the letter →

arxiv 1909.01881 v1 pith:2G2RJIVS submitted 2019-09-04 math.AP

Ruipeng Shen This is my paper · ORCID

classification math.AP MSC 35L7135L05

keywords defocusingnonlinearwaveequationradialsolutionsscatteringinwardenergyoutwardweightedMorawetzestimateLpL2pspacetimenormdistribution

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper proves a new criterion for scattering of radial, finite-energy solutions of the three-dimensional defocusing semilinear wave equation $\partial_t^2 u - \Delta u = -|u|^{p-1}u$ with $3 \leq p < 5$. The criterion is one-sided: only the inward part of the energy at time zero must decay, with weight $\max\{1,|x|^\kappa\}$ for $\kappa \geq (5-p)/(p+1)$, while the outward energy may be arbitrarily large and slowly decaying. Scattering means the solution approaches a free solution of the linear wave equation in $\dot H^1 \times L^2$ as $t \to +\infty$. When $\kappa$ is strictly above the threshold, the paper additionally gives a power-law rate for the spacetime norm $\|u\|_{L^p L^{2p}(\mathbb{R}_+ \times \mathbb{R}^3)}$ and for the convergence rate, tied directly to the excess decay $\kappa - \kappa_0(p)$. The significance is that finite energy alone has not been enough in this subcritical range, and the new assumption concerns only the energy moving inward, which matches the physical intuition that inward-moving energy is the only possible obstruction to forward scattering.

What carries the argument

The argument runs through the radial reduction $w(r,t) = r u(|x|,t)$, which converts the 3D equation into a 1D wave equation with nonlinearity $-|w|^{p-1}w/r^{p-1}$ and preserves energy up to a constant. In this reduction, inward and outward energies are $E_\mp(t) = \pi \int_0^\infty (|w_r \pm w_t|^2 + \frac{2}{p+1}|w|^{p+1}/r^{p-1})\, dr$. The load-bearing structural object is the energy flux formula, a boundary-integral identity with a distinguished $L^2$ function $\xi(t)$ that records energy transferred from inward to outward motion through the origin; together with the triangle laws it expresses local energies as sums of flux integrals and spacetime integrals of $|w|^{p+1}/r^p$. Multiplying the flux identity by a weight $a(r+t)$ that grows like $(r+t)^\kappa$ yields the weighted Morawetz estimates $\int_0^\infty t^\kappa |\xi(t)|^2\, dt \lesssim K$ and $\int\!\int (r+t)^\kappa |w|^{p+1}/r^p\, dr\, dt \lesssim K$, which imply $E_-(t) \lesssim K t^{-\kappa}$ and the energy-in-cylinder bounds used later. The scattering proof then uses a prior energy-location theorem saying non-scattering energy must persist in an annulus just inside the light cone, and shows the weighted estimates make that annulus empty in the limit.

What would settle it

For a fixed $p \in [3,5)$ and $\kappa \geq \kappa_0$, produce one radial finite-energy solution satisfying the inward decay condition whose $\dot H^1 \times L^2$ distance to every free wave fails to tend to zero; or, for the rate part, compute $\|u\|^p_{L^p L^{2p}([t,\infty))}$ on a high-resolution radial simulation and find decay slower than $C t^{-\frac{p+1}{p+3}(\kappa-\kappa_0)}$. A single such example would disprove Theorem 1.3.

Watch

Extended reading notes

Core claim

The central claim, Theorem 1.3, is that radial initial data with finite energy scatter in positive time provided the inward energy satisfies $\int_{\mathbb{R}^3} \max\{1,|x|^\kappa\}(|\nabla u_0\cdot x/|x| + u_0/|x| + u_1|^2 + \frac{2}{p+1}|u_0|^{p+1})\, dx < \infty$ with $\kappa \geq \kappa_0(p) = (5-p)/(p+1)$. Here the integrand is the inward energy density, the combination of radial momentum and potential energy moving toward the origin. The proof shows that this decay of inward energy forces $E_-(t) \leq C K t^{-\kappa}$ for all later times via a weighted Morawetz estimate, and then combines this decay with a previously established energy-distribution theorem: if a solution failed to scatter, a fixed positive amount of energy would have to remain in the annulus $t/2 < |x| < t - t^\beta$ for every $\beta < 2(p-2)/(p+1)$. For $\kappa > \kappa_0$, the decay and annulus bounds contradict that persistence directly; at the endpoint $\kappa = \kappa_0$, an additional $L^2$ convergence statement along lightlike characteristics (Proposition 2.11) removes the remaining gap and still yields a contradiction. For $\kappa > \kappa_0$ the same estimates prove $\|u\|_{L^p L^{2p}}^p$ on $[t,\infty)$ is bounded by $C t^{-\frac{p+1}{p+3}(\kappa-\kappa_0)}$, and the distance in $\dot H^1 \times L^2$ from the evolved solution to the asymptotic free wave obeys the same bound.

Load-bearing premise

The contradiction step assumes as a black box the earlier theorem that a non-scattering radial solution must keep positive energy in the annulus $t/2 < |x| < t - t^\beta$ for every $\beta < 2(p-2)/(p+1)$; if that energy-location result fails or does not apply forward in time, the scattering conclusion collapses.

Editorial extensions

If this is right

If the inward energy decays at the endpoint rate $\kappa_0 = (5-p)/(p+1)$, every radial finite-energy solution scatters in forward time; no assumption on the size or decay of outward energy is needed beyond finiteness of total energy.
For any $\kappa > \kappa_0$, the nonlinear forcing satisfies $\| |u|^{p-1}u \|_{L^1 L^2([t,\infty)\times \mathbb{R}^3)} \leq C t^{-\frac{p+1}{p+3}(\kappa-\kappa_0)}$, so the solution converges to a free wave at that explicit polynomial rate.
By time reversal, the same statement holds backward in time if the outward energy decays with the same weight and the sign of the initial velocity is flipped.
The appendix constructs a scattering solution whose $L^{2(p-1)} L^{2(p-1)}$ norm in the critical-Sobolev sense is infinite, so the new criterion covers scattering phenomena invisible to the critical-space scattering theory.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the energy-location theorem could be sharpened below $\kappa_0$, the same weighted Morawetz estimates would likely push the scattering threshold lower: for $3 < p < 5$ the paper's own Remark 4.5 already shows the medium-radius contribution decays when $\kappa$ is slightly below $\kappa_0$.
The mechanism isolates the light-cone annulus $t/2 < |x| < t - t^\beta$ as the only possible obstruction to scattering, so a numerical test of near-threshold radial data should look specifically at whether energy accumulates there.
The flux identity with a growing weight is not obviously specific to the power nonlinearity; similar inward-energy decay conditions may imply scattering for other defocusing nonlinearities or other dimensions where a radial 1D reduction and a $\xi$-flux identity are available.
The explicit rate suggests an optimal-scattering-rate question: whether the exponent $\frac{p+1}{p+3}(\kappa-\kappa_0)$ is the true worst-case decay for this class, or merely an upper bound.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Ruipeng has a genuinely new scattering criterion here: for radial, energy-subcritical defocusing NLW in 3D, you only need the inward energy to decay at rate |x|^kappa with kappa >= (5-p)/(p+1), and you get forward scattering, including the endpoint kappa0 that was previously open. The paper also produces explicit decay for the L^p L^{2p} norm and for the convergence rate when kappa > kappa0. That is a real step beyond the earlier work in [12], which assumed weighted decay of the total energy.

The weighted Morawetz estimates in Section 3 looked sound to me; they are worked out in detail and the decay corollaries follow. The decomposition in Section 4.2 for the L^p L^{2p} bound is careful, and the pointwise estimate of Lemma 2.4 does a lot of work. The endpoint case uses the g+ approximation from [12] in a clean way.

The soft spot, and the one I would want addressed before betting on Theorem 1.3, is the contradiction argument in Section 4.1. It opens by invoking part (c) of Theorem 1.2 to say that if the solution does not scatter, energy remains in the annulus [t/2, t - t^beta] as t goes to infinity. But the manuscript's Theorem 1.2 states the energy-location result only for negative times; for positive times it just says 'the asymptotic behaviour is similar.' The proof needs that positive-time version, and it is neither stated precisely nor reproved. This is not circular--the inward-energy decay hypothesis is used earlier, in the weighted Morawetz--but it makes Theorem 1.3(a) conditional on a statement that lives in a separate preprint [12]. If [12] is solid, the rest of the scattering proof hangs together. If not, the whole thing falls. The author should fix this before publication, either by including the exact statement and proof of the positive-time analogue or by clearly citing a published version.

Minor: the appendix example is interesting but a bit tangential; it demonstrates a limitation of critical-space scattering theory, not a step in the main proof. Not a problem, just a note.

This deserves a serious referee. The central result is new, the estimates are mostly self-contained, and the dependency on [12] is explicit enough to be checked. I would send it out and ask the author to resolve the positive-time Theorem 1.2(c) issue in revision. If I were working in scattering theory for NLW, I would cite it.

Referee Report

2 major / 3 minor

Summary. This paper studies the 3D defocusing energy-subcritical nonlinear wave equation ∂_t^2 u − Δu = −|u|^{p−1}u for 3≤p<5 in the radial finite-energy setting. The main result (Theorem 1.3) asserts that if the inward component of the initial energy satisfies a weighted decay condition with weight max{1,|x|^κ} and κ≥κ0(p)=(5−p)/(p+1), then the solution scatters forward in Ḣ^1×L^2; for κ>κ0(p), the solution also has finite L^pL^{2p}(R_+×R^3) norm, with explicit decay rates for the nonlinearity and for convergence to a free wave. The proof combines a new weighted Morawetz estimate (Section 3), giving decay of the inward energy and of the local outward energy, with a contradiction argument (Section 4.1) based on energy-distribution results imported from the author's earlier work [12]. The appendix constructs an example of a scattering solution with infinite L^{2(p−1)}L^{2(p−1)} norm, showing the phenomenon lies outside the critical-space scattering theory.

Significance. If correct, the result is a substantial advance: it replaces the usual assumptions of full weighted energy decay or compact support by a one-sided (inward) energy decay condition, it covers the endpoint κ=κ0, and it supplies quantitative rates for the L^pL^{2p} norm and the approach to the free wave. The weighted Morawetz estimates in Section 3 are derived in detail and appear internally consistent, and the geometric decomposition in Section 4.2 is coherent. The main caveat is the paper's reliance on a large body of unproved results from the author's preprint [12], in particular on a forward-time version of the energy-location theorem that is not stated in the manuscript; the central claim is therefore conditional on those external inputs.

major comments (2)

[§4.1, equation (11)] The contradiction step assumes, with reference to the third bullet of Theorem 1.2, that if the solution did not scatter forward then lim_{t→∞} E(t;t/2,t−t^β)>0 for every β<2(p−2)/(p+1). However, Theorem 1.2 as stated in this paper proves the energy-location statement only in the negative time direction; the positive time direction is dismissed with the sentence 'The asymptotic behaviour in the positive time direction is similar.' Since Theorem 1.3(a) rests on this forward-time dichotomy, the manuscript must state the forward-time version explicitly and either prove it (for instance by applying Theorem 1.2 to the time-reversed solution, which is legitimate because Theorem 1.2 concerns all radial finite-energy solutions) or supply a precise theorem number in [12]. Without this, the scattering proof is conditional on an unstated external result.
[§2.3, Propositions 2.5–2.11] The paper imports substantial machinery from the author's preprint [12] without proof: the general energy flux formula involving the function ξ (Proposition 2.5), the triangle law (Proposition 2.7), the identity for E_-(t) (Proposition 2.8), the monotonicity and flux limits (Propositions 2.9–2.10), and the convergence of the characteristic traces (Proposition 2.11). These results are used directly in the proof of Theorem 1.3, yet [12] is an arXiv preprint and no proofs are given here. The manuscript should either include proofs of these statements or provide precise theorem numbers in [12] so that the reader can verify them; in its present form the central theorem is not self-contained.

minor comments (3)

[§4.1, endpoint κ=κ0] In the displayed estimate for the endpoint case, the change of variables r=t−τ gives τ=t−r, so the interval r∈[t−ct^{1−κ0}, t−t^β] corresponds to τ∈[t^β, ct^{1−κ0}], not τ∈[ct^{1−κ0}, t^β]. As written the integrals on the right have reversed limits and are negative for large t; the intended argument is clear, but the formulas should be corrected.
[Abstract and Theorem 1.3] The abstract states the condition as κ≥κ0 without the upper bound κ<1, while Theorem 1.3 requires κ∈[κ0,1). The abstract (and the informal description in Section 1.2) should state the restriction κ<1 explicitly.
[Section 1.3] In the definition of the one-dimensional energy E(w,w_t), the term 'r^{r−1}' is a typo; it should be 'r^{p−1}'.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the proof derives scattering from a weighted Morawetz estimate plus an independently stated energy-location theorem; reliance on the author's earlier preprint is a dependency, not a circular reduction.

full rationale

The derivation chain is not circular. The weighted Morawetz estimates in Proposition 3.1 are obtained from the energy-flux identity (Proposition 2.5, imported from [12]) by multiplying by a'(s), integrating over s, and applying the triangle law; the resulting decay estimates E_-(t) <= C K t^{-kappa}, Proposition 3.4, and Proposition 3.5 are consequences of those estimates, not restatements of the hypothesis. The scattering contradiction in Section 4.1 uses the positive-time analogue of Theorem 1.2(c) from [12] to assert the limit in (11) for a hypothetical non-scattering solution. That positive-time statement is indeed only asserted here via the sentence 'The asymptotic behaviour in the positive time direction is similar,' so the proof is not fully self-contained; however, this is an external dependency, not circularity. The cited theorem's assumptions (finite-energy radial solutions, 3 <= p < 5) do not include the inward-energy-decay hypothesis whose consequences are being derived, and the paper never defines the scattering state or the claimed convergence rate from the assumed quantity K. No equation is used to define its own conclusion, and no fitted parameter is relabelled as a prediction.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No numerical parameters are fitted, and the theorem has no invented physical or mathematical entities. The inward and outward energies are definitions carried over from the author's prior work and are analytical bookkeeping, not new postulated objects. The main external input is the earlier inward/outward energy machinery, especially Theorem 1.2 from [12], which is a dependency on prior work rather than an invented entity.

assumptions (4)

domain assumption Theorem 1.2 and Propositions 2.5 through 2.11 of [12] (energy distribution, triangle laws, monotonicity, characteristic convergence) are valid.
Invoked throughout Section 2.3 and used decisively in Section 4.1 to turn failure of scattering into the positive energy limit (11). Not proved in this paper.
standard math Strichartz estimates and local well-posedness for the 3D wave equation hold as stated in [5], as used in Lemma 4.2.
Cited to Ginibre and Velo and used in Section 4.2 to control local pieces of the solution.
standard math The Kenig-Merle radial pointwise bound |u(r)| <= C r^{-1/2} ||u||_{dot H^1} and the improved pointwise estimate in Lemma 2.4 are valid.
Lemma 2.3 is cited to [8]; Lemma 2.4 is proved in the text and used for the L^p L^{2p} bounds.
standard math The classic Morawetz estimate gives integral of |u|^{p+1}/|x| over space-time bounded by energy, making the flux identities well defined.
Remark 2.6 uses this to justify finiteness of double integrals in the energy flux formulas.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Scattering of solutions to NLW by Inward Energy Decay." pith.science (2026). https://pith.science/paper/2G2RJIVS

@misc{pith2026190901881,
  author       = {Pith},
  title        = {Pith review of: Scattering of solutions to NLW by Inward Energy Decay},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2G2RJIVS}},
  note         = {Machine review of arXiv:1909.01881}
}

abstract

The topic of this paper is a semi-linear, energy sub-critical, defocusing wave equation $\partial_t^2 u - \Delta u = - |u|^{p -1} u$ in the 3-dimensional space ($3\leq p<5$) whose initial data are radial and come with a finite energy. In this work we prove scattering in the positive time direction by only assuming the inward part of the energy decays at a certain rate, as long as the total energy is finite, regardless of the decay rate or size of the outward energy. More precisely, we assume the initial data comes with a finite energy and \[ \int_{{\mathbb R}^3} \max\{1,|x|^\kappa\}\ (\ |\nabla u_0(x)\cdot \frac{x}{|x|} + \frac{u_0(x)}{|x|} + u_1(x)\ |^2 + \frac{2}{p+1}|u_0(x)|^{p+1}\ ) dx < \infty. \] Here $\kappa\geq \kappa_0(p) = \frac{5-p}{p+1}$ is a constant. If $\kappa>\kappa_0(p)$, we can also prove $\|u\|_{L^p L^{2p}}({\mathbb R}^+ \times {\mathbb R}^3)< +\infty$ and give an explicit rate of $u$'s convergence to a free wave.

Figures

Figures reproduced from arXiv: 1909.01881 by the authors.

**Figure 2.** Illustration of proof for Proposition 3.5 [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗

**Figure 3.** Illustration of proof for Proposition 3.5 [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Illustration of proof for Lemma 5.1 with (wr(·, t), wt(·, t)) ∈ C(R;L 2 (R +) × L 2 (R +)). In addition, the initial data (w0, w1) satisfy |w0(r)| ≤ crβ , r ≥ R; [PITH_FULL_IMAGE:figures/full_fig_p022_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 12 canonical work pages

[12]

Energy Distribution of Radial Solutions to Energy Subcritical Wave Equation with an Application on Scattering Theory

R. Shen. “Energy Distribution of Radial Solutions to Energy Subcritical Wave Equation with an Application on Scattering Theory” arXiv preprint 1808.08656. 24

work page Pith review arXiv
[1]

Global well-posedness and scattering for the radial, defocusing, cubic nonlinear wave equation

B. Dodson. “Global well-posedness and scattering for the radial, defocusing, cubic nonlinear wave equation” arXiv preprint 1809.08284

work page Pith review arXiv
[2]

Scattering for the radial 3d cubic wave equation

B. Dodson and A. Lawrie. “Scattering for the radial 3d cubic wave equation.” Analysis and PDE, 8(2015): 467-497

work page 2015
[3]

Scattering for radial, bounded solutions of focusing supercritical wave equations

T. Duyckaerts, C.E. Kenig, and F. Merle. “Scattering for radial, bounded solutions of focusing supercritical wave equations.” International Mathematics Research Notices 2014: 224-258

work page 2014
[4]

Conformal invariance and time decay for nonlinear wave equa- tions

J. Ginibre, and G. Velo. “Conformal invariance and time decay for nonlinear wave equa- tions.” Annales de l’institut Henri Poincar´ e (A) Physique th´ eorique47(1987), 221-276

work page 1987
[5]

Generalized Strichartz inequality for the wave equation

J. Ginibre, and G. Velo. “Generalized Strichartz inequality for the wave equation.” Journal of Functional Analysis 133(1995): 50-68. 23

work page 1995
[6]

Regularity and asymptotic behaviour of the wave equation with critical non- linearity

M. Grillakis. “Regularity and asymptotic behaviour of the wave equation with critical non- linearity.” Annals of Mathematics 132(1990): 485-509

work page 1990
[7]

Conformal conservation law, time decay and scattering for nonlinear wave equation

K. Hidano. “Conformal conservation law, time decay and scattering for nonlinear wave equation” Journal D’analysis Math´ ematique91(2003): 269-295

work page 2003

Show all 12 references

[8]

Nondispersive radial solutions to energy supercritical non-linear wave equations, with applications

C. E. Kenig, and F. Merle. “Nondispersive radial solutions to energy supercritical non-linear wave equations, with applications.” American Journal of Mathematics 133, No 4(2011): 1029-1065

2011
[9]

On existence and scattering with minimal regularity for semi- linear wave equations

H. Lindblad, and C. Sogge. “On existence and scattering with minimal regularity for semi- linear wave equations” Journal of Functional Analysis 130(1995): 357-426

1995
[10]

Morrey-Campanato estimates for Helmholtz equations

B. Perthame, and L. Vega. “Morrey-Campanato estimates for Helmholtz equations.”Journal of Functional Analysis 164(1999): 340-355

1999
[11]

On the energy subcritical, nonlinear wave equation in R3 with radial data

R. Shen. “On the energy subcritical, nonlinear wave equation in R3 with radial data” Anal- ysis and PDE 6(2013): 1929-1987

2013

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [12]

Energy Distribution of Radial Solutions to Energy Subcritical Wave Equation with an Application on Scattering Theory

R. Shen. “Energy Distribution of Radial Solutions to Energy Subcritical Wave Equation with an Application on Scattering Theory” arXiv preprint 1808.08656. 24

work page Pith review arXiv

[2] [1]

Global well-posedness and scattering for the radial, defocusing, cubic nonlinear wave equation

B. Dodson. “Global well-posedness and scattering for the radial, defocusing, cubic nonlinear wave equation” arXiv preprint 1809.08284

work page Pith review arXiv

[3] [2]

Scattering for the radial 3d cubic wave equation

B. Dodson and A. Lawrie. “Scattering for the radial 3d cubic wave equation.” Analysis and PDE, 8(2015): 467-497

work page 2015

[4] [3]

Scattering for radial, bounded solutions of focusing supercritical wave equations

T. Duyckaerts, C.E. Kenig, and F. Merle. “Scattering for radial, bounded solutions of focusing supercritical wave equations.” International Mathematics Research Notices 2014: 224-258

work page 2014

[5] [4]

Conformal invariance and time decay for nonlinear wave equa- tions

J. Ginibre, and G. Velo. “Conformal invariance and time decay for nonlinear wave equa- tions.” Annales de l’institut Henri Poincar´ e (A) Physique th´ eorique47(1987), 221-276

work page 1987

[6] [5]

Generalized Strichartz inequality for the wave equation

J. Ginibre, and G. Velo. “Generalized Strichartz inequality for the wave equation.” Journal of Functional Analysis 133(1995): 50-68. 23

work page 1995

[7] [6]

Regularity and asymptotic behaviour of the wave equation with critical non- linearity

M. Grillakis. “Regularity and asymptotic behaviour of the wave equation with critical non- linearity.” Annals of Mathematics 132(1990): 485-509

work page 1990

[8] [7]

Conformal conservation law, time decay and scattering for nonlinear wave equation

K. Hidano. “Conformal conservation law, time decay and scattering for nonlinear wave equation” Journal D’analysis Math´ ematique91(2003): 269-295

work page 2003