Skip to content

CHANGELOG.md

Latest commit

 

History

History
43 lines (29 loc) · 3.9 KB

CHANGELOG.md

File metadata and controls

43 lines (29 loc) · 3.9 KB
Raw

Changelog

All notable changes to paper-tle-divergence-atlas are documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[0.2.0] — 2026-05-24

Reproducibility refresh of v0.1.0 under the gmat-sweep / gmat-run 0.6 toolchain. Per-row Δr against the v0.1.0 outputs is at floating-point-roundoff scale across every aggregate, summary, and diagnostic — no manuscript reissue, no arXiv v2, no science changes.

Changed

  • Sweep, CdA sensitivity arm, maneuver-threshold sensitivity arm, and sweep/manifest.jsonl regenerated end-to-end against gmat-sweep / gmat-run 0.6 (#68). The refreshed Zenodo deposit is published as version 0.2.0 under the existing concept DOI 10.5281/zenodo.20277028 (version DOI 10.5281/zenodo.20370326).
  • Pipeline upgraded to gmat-sweep / gmat-run 0.6 (#62), dropping the 0.4-era script-templating workaround.
  • Phase-3 post-processing folded into the sweep via gmat-sweep's postprocess hook (#63); the hook is now re-applied on Sweep resume so partial-restart runs match cold-start runs bit-for-bit (#66).

Added

  • arXiv preprint badge and citation metadata pointing at arXiv:2605.19850 (#58).
  • Regression test pinning run_id preservation when aggregating the CdA sensitivity arm (#67).

Removed

  • make figures Makefile target (#71); figures are produced by make build via showyourwork's Snakemake DAG.

[0.1.0] — 2026-05-18

Initial citeable release of How long can you trust a Starlink TLE? An empirical comparison of SGP4 and high-fidelity propagation against operator-updated truth across a megaconstellation.

Added

  • 24,641-pair locked sweep corpus across 501 Starlink satellites, stratified by altitude shell (540 / 550 / 560 km) and platform generation (v1.0, v1.5, v2-mini) over April 2026, with per-pair SGP4 and GMAT high-fidelity propagation evaluated against operator next-TLE proxy truth.
  • High-fidelity force model: EGM2008 70×70 + Sun/Moon point-mass third bodies + NRLMSISE-00 drag + conical-shadow SRP, integrated with Runge–Kutta 8(9) to 10⁻¹² km tolerance.
  • Truth-floor diagnostic (§2.1, Table 1) framing the 6-hour headline as floor-limited at ~1 km median against the next-TLE proxy.
  • Dynamical-consistency caveat (§3.6.1) acknowledging the Brouwer mean-element mismatch between the SGP4 initial state and the high-fidelity propagator.
  • Statistical-estimator section (§3.7.1): per-pair power-law fit on log-log with satellite-level bootstrap CIs, likelihood-ratio tests against the k = 1 / k = 2 nulls, and a mixed-effects parametric cross-check in Appendix B.4.
  • H3 regression specification (§3.7.2): per-shell ANCOVA fit of the per-satellite SGP4 staleness coefficient against daily-observed F10.7, with the 81-day-centred robustness check.
  • Sensitivity studies bundled with the main sweep: CdA × {0.8, 1.2} on the v2-mini cohort (Appendix B.2) and maneuver-threshold ±50 m / +200 m perturbations (Appendix B.3).
  • §5.2 cohort-resolved view of the v2-mini long-Δt H2 reading, with Table 8 reporting per-(shell × generation × Δt) win fractions and quantifying the v2-mini majority-wins regime at 7 d across both populated shells.
  • Per-cell power-law (A, k) atlas in Table 3, intended for downstream use as a benchmark target for enhanced-propagator work (SGP4-XP, differentiable SGP4, ML-residual correctors).
  • make bundle target producing the canonical 17-file Zenodo deposit at the repository root.
  • make arxiv-tarball target wrapping showyourwork tarball with a post-processor that strips dotfiles and showyourwork v0.4.3's root-level figure/table duplicates.
  • Reproducibility surface: make build renders ms.pdf from a Zenodo-cached sweep bundle on a clean checkout with no local GMAT installation; make sweep reproduces the bundle from scratch given GMAT R2026a.