Editing Openai/6897769e-4ee4-800f-aba5-69cca34f701c (section)

=== • Dispersion model: the Cauchy-style n(λ) I used is a rough approximation. Real water dispersion (Sellmeier data or measured tabulated n(λ)) is needed to get an accurate δmin(λ) vs λ curve. Small changes in n(λ) produce degree-level changes in δmin. ===

• Spectral weighting: I used a Sun-like Planck spectrum as one plausible example. Different sources (stellar, black-body at other T, atomic emission spectra, line-dominated plasmas) will change the weighted mean dramatically.

• Geometry choice: I concentrated on the single-reflection primary rainbow deflection because it is an explicit spherical-light scattering formula and it makes the spherical droplet link you mentioned (rainbow 42° etc.) direct. But other spherical processes (internal scattering from droplets, diffraction from spheres, electron-photon interaction geometry around atoms) will have different angle functions and may produce smaller offsets.

• Statistical vs exact: if the QAT mechanism is inherently statistical (broken spherical symmetry, imperfect geometry), expecting a simple exact numerical proof like a Euclidean theorem may be too strong. The small offset (≈0.47°) might be a statistical/ensemble effect rather than a pure geometrical identity.