A Unified Integral Equation Approach to Conservation Laws for Nonlinear Schrödinger Equations

Technical Summary: A Unified Integral Equation Approach to Conservation Laws for Nonlinear Schrödinger Equations

Problem Statement
The paper addresses the rigorous derivation of conservation laws and related identities for the Cauchy problem of the nonlinear Schrödinger equation (NLS) with power-type nonlinearities. The equation is given by:
$$i\partial_t u + \frac{1}{2}\Delta u = f(u), \quad u(0, x) = \phi(x)$$
where $f$ satisfies specific analytic and algebraic conditions (A1–A3) ensuring local well-posedness and gauge invariance.

Traditionally, conservation laws such as charge (mass), energy, momentum, the pseudo-conformal conservation law, and virial identities are derived formally by pairing the differential equation with specific test functions (e.g., $iu$, $-\partial_t u$, $\nabla u$). However, these formal computations require high regularity (typically $H^2$ or $H^2 \cap \mathcal{F}H^2$) to justify the integration by parts and divergence theorem steps. For solutions with lower regularity (e.g., $L^2$, $H^s$, $H^1$, or $\Sigma = H^1 \cap \mathcal{F}H^1$), these formal derivations are not directly applicable.

Standard rigorous justifications rely on two approximation methods:

1. Approximating the solution by smooth solutions and using continuous dependence on initial data.
2. Regularizing the equation (e.g., Yosida approximation) or introducing spatial weights (e.g., $e^{-\epsilon|x|^2}$) to recover integrability, followed by delicate limiting arguments.

The paper notes that the justification of conservation laws is closely linked to the uniqueness theory of the solution class, as uniqueness implies energy conservation via time-reversal invariance.

Methodology
The authors propose a unified framework that derives conservation laws directly from the integral equation (Duhamel formulation) associated with the NLS, without employing smooth approximations or regularization procedures. The approach relies on:

• Strichartz Estimates: Utilizing the space-time integrability provided by these estimates to define solutions as mild solutions in appropriate Banach spaces (e.g., $C(I; H^s) \cap L^\rho(I; H^{s,\gamma})$).
• Master Integral Identity: The core of the methodology is a single integral identity (Proposition 3.1) governing the inner product of two solutions to integral equations. This identity serves as a rigorous counterpart to the formal pairing identities at the differential equation level.
• Duality Pairings: The framework operates within a specific functional setting involving Lebesgue and Sobolev spaces (specifically pairs like $(L^2 \cap L^r) \times (L^2 + L^{r'})$) to ensure all duality pairings and time integrals are well-defined for low-regularity solutions.

Key Contributions and Results
The paper establishes a "Master Identity" (Proposition 3.1) which states that for solutions $v_1, v_2$ to integral equations driven by source terms $g_1, g_2$, the inner product evolves as:
$$(v_1(t), v_2(t)) = (\psi_1, \psi_2) + i \int_0^t \left( \langle v_1(s), g_2(s) \rangle - \langle v_2(s), g_1(s) \rangle \right) ds$$
From this single identity, the authors systematically derive the following conservation laws and identities for solutions with minimal regularity requirements:

1. Conservation of Charge (Mass): Derived for $L^2$-solutions by setting $v_1=v_2=u$ and $g=f(u)$. The vanishing of the nonlinear term follows from the algebraic condition $\text{Im}(\bar{z}f(z)) = 0$.
2. Conservation of Energy: Derived for $H^1$-solutions by applying the gradient operator to the integral equation and utilizing the Hamiltonian structure of the nonlinearity.
3. Conservation of Momentum: Derived for $H^1$-solutions by pairing the integral equation for $u$ with that of $\nabla u$. The proof relies on the cancellation property of the nonlinear term under the divergence theorem, justified via duality pairings.
4. Pseudo-Conformal Conservation Law: Derived for $\Sigma$-solutions (where $x\phi \in L^2$) by utilizing the operator $J(t) = x + it\nabla$. The derivation avoids the need for the pseudo-conformal conservation law to prove virial identities, unlike previous approaches.
5. Virial-Type Identities:
   • A first-order virial identity is derived directly from the master identity applied to $xu$, avoiding the need for spatial weight regularization.
   • A cross-term identity relating $\text{Im}(\nabla u, xu)$ to the energy and the potential scaling term $W(u)$ is established.
   • The standard virial identity is obtained by combining the first-order identity with the cross-term identity and energy conservation.

Significance and Claims
The paper claims that this approach offers a "unified framework" where all major conservation laws and identities follow systematically from a single integral identity. The primary significance lies in the elimination of approximation sequences and auxiliary regularization. By working entirely at the level of the integral equation and leveraging Strichartz estimates, the authors provide a rigorous justification for these laws that applies directly to the mild solutions defined by the standard well-posedness theory.

The authors note that while the specific setting is the NLS, the underlying mechanism is abstract and relies on a robust dual-pairing framework compatible with dispersive estimates. They suggest that this method could potentially extend to broader classes of dispersive or wave equations, particularly in settings where approximation-based derivations are delicate, such as those involving singular interactions or quantum hydrodynamic models. However, the paper does not claim to have solved these broader problems, but rather identifies the integral-equation approach as a relevant tool for such investigations.