Boundary Information Geometry B9: Metastability Hierarchies and an Empirical Structural Comparison with Nuclear Fission

This preprint reports BIG-B9 empirical phase II, a finite two-dimensional reduced-model study of metastability generated by boundary cost and repulsive nonlocal interaction. The main result is a hierarchy of distinct model scales: far-separated energetic preference near lambda~42-44, near-scission preference at lambda=55.917826, smooth m=2 finite-amplitude coexistence at lambda=89.194319, smooth m=2 local spinodal at lambda=89.739090, and a diagnostic p=1 pinch-envelope zero near lambda~102.7. The coexistence and spinodal scales were reproduced in a prospectively frozen held-out-resolution test. A descriptive fissility normalization places 52 RIPL-3 nuclei with 88<=Z<=96 in the model regime where separated configurations are energetically favored while the compact reference remains locally stable. The comparison is structural, not a calibrated nuclear-fission calculation. A limited two-case external barrier-magnitude check was mixed, two prospective near-spinodal exponent tests failed, and the quartic coefficient remains numerically unresolved. The release preserves those negative results and does not claim a quantitative lambda-to-MeV mapping, a resolved quadratic/cubic barrier law, or reduction of shell, pairing, tunneling, and other microscopic nuclear effects to the present boundary variables. Project repository For the current research map, related BIG studies, representative figures, terminology, and links to the broader Boundary Information Geometry (BIG) research program, see: Boundary Information Geometry (BIG) — GitHub Repositoryhttps://github.com/Jun-Lucis/BIG-theory The Zenodo record remains the archival and citable source for this research output. The GitHub repository serves as a navigational overview of the continuing BIG research program.