MEON R56 STScI Validation: Team-Normalized Topological Excess in Hubble Frontier Fields Lens-Model Reconstructions

This Zenodo deposit documents the R56 external validation stage of the MEON framework, testing whether topological structure metrics derived from real Hubble Frontier Fields data remain distinguishable from external STScI lens-model FITS reconstructions. The validation chain includes four linked audits: R56-FIX3, a Williams-only external FITS comparison yielding CONDITIONAL_PASS; R56G, a raw multi-team STScI comparison showing FAIL under unnormalized pooling due to team- and scale-dependent reconstruction effects; R56H, a team-normalized multi-team audit yielding PASS for the core topology metrics; and R56I, a leave-one-cluster-out team-normalized robustness audit yielding PASS across all six held-out HFF clusters.

The core result is that the MEON topology signal is not supported by raw multi-team pooling, but reappears after robust team-internal normalization and survives cluster-held-out validation. The strongest and most stable signal is a lacunarity-dominated topological excess, supported by component-structure evidence. Scale-sensitive diagnostics such as alpha_energy_proxy and mu_proxy_p95 are reported separately and are not used as primary physical decision metrics.

This deposit includes the PDF report, Python audit scripts, result tables, plots, and summary files required to reproduce and inspect the R56 validation chain. The results should be interpreted as evidence for a robust residual topological contrast between real HFF-derived maps and external STScI multi-team lens-model reconstructions. They should not be interpreted as a hydrodynamical LambdaCDM raytracing test, an Ares/Hera simulation comparison, or a direct falsification of dark matter or LambdaCDM.