High-Frequency Instabilities of the Kawahara Equation: A Perturbative Approach

The paper studies high-frequency instabilities of small-amplitude Stokes waves for the Kawahara equation and derives, for the two principal resonance cases Δn=1 and Δn=2, explicit leading-order formulas for: (i) the interval of Floquet exponents μ that parameterize the isola, (ii) the most unstable spectral element, and (iii) an ellipse asymptotic to the isola. For Δn=1, the paper gives M1, the μ-interval (μ0 ± ε M1), the ellipse (4.16), and λ*,r = ε |σ|√−(μ0+m)(μ0+n) with μ* = μ0 + O(ε^2) (see 4.11–4.17) . For Δn=2, it introduces C_N^{μ2,μ0}, P_N^{μ0}, and S2, solves a 2×2 solvability system to obtain the det-formula for λ2 (4.28a), the μ-interval (4.29)–(4.30), the most unstable point (4.31)–(4.32), and the ellipse (4.33), together with the nonvanishing of S2 (Corollary 2) . The candidate solution reproduces these results via a two-mode Floquet–Bloch/Lyapunov–Schmidt (Schur-complement) reduction: for Δn=1 it builds a leading-order 2×2 dispersion matrix with O(ε) couplings and recovers exactly the paper’s M1, ellipse (with the same axis ratio), and λ*,r scaling; for Δn=2 it includes the order-ε^2 direct second-harmonic and virtual transitions through the intermediate mode, obtains the same reduced 2×2 determinant with C_N^{μ2,μ0}, P_N^{μ0}, and S2, the same μ-interval, most-unstable point, and ellipse (matching 4.28–4.34). The background operator, dispersion relation, and Floquet formulation used by both coincide with the paper’s setup (3.3)–(3.8), (2.13)–(2.14) . Minor differences are present in presentation and explicit listing of assumptions: the paper states nonresonance and Krein-sign constraints and (for Δn=2) parity/evenness needed at O(ε^2), while the candidate assumes a generic nonresonant setting and does not spell out the evenness of u2. Overall, the conclusions agree and the proofs, although closely related in spirit, are presented through different reductions.