From Reconstruction to Prediction: Scale-Dependent χ-Field Closure for Galaxy Rotation Curves

This repository contains the complete experimental artifacts supporting the paper: “From Reconstruction to Prediction: Scale-Dependent χ-Field Closure for Galaxy Rotation Curves”Greg D. Partin, LFM Research (2026) The study investigates whether χ-fields governed by Klein–Gordon–type dynamics can predict galaxy rotation curves beyond the region used for reconstruction. Using 91 galaxies from the SPARC database, χ is reconstructed from the inner 70% of each galaxy’s radial extent, and rotation velocities are predicted in the outer 30% using multiple extrapolation closures under a strict train/test protocol. Six closure variants (A1–A6) are evaluated. All closures using the physical speed of light in the χ→velocity mapping fail catastrophically, producing relativistic velocities. Prediction becomes viable only after identifying an effective galactic-scale propagation speed ceff=300c_{\mathrm{eff}} = 300ceff=300 km/s, applied globally with no per-galaxy tuning. With this single constant, the exponential-tail closure (A6) achieves reduced chi-squared χ²ν ≈ 1, consistent with the observational uncertainty floor. This archive includes: All source code used to acquire, process, and analyze SPARC data Configuration files and environment specifications Processed datasets Per-galaxy and summary result tables (CSV/JSON) Publication-quality figures (PNG/PDF) Full documentation enabling deterministic reproduction of results All experiments are deterministic and rely exclusively on publicly available data. No proprietary software or non-public inputs are required. The contents of this repository are intended to support transparency, reproducibility, and independent verification of the results reported in the associated manuscript.