Skip to content

Solar systems and galaxies: where Kepler's law holds, where rotation curves go flat, and the little solar system at the Galaxy's centre #36

Open

Description

@ahrazzle

Question

The README opens with "atoms look a little like solar systems, solar systems a little like galaxies". The first half of that line has a study (the Witness Map). The second half has none.

A solar system and a spiral galaxy are both flattened, rotating disks, but they orbit differently:

  • The Solar System. The Sun holds about 99.86% of the mass, so orbital speeds fall as Kepler's v ∝ r^−1/2.
  • The Milky Way. The circular speed stays near 230 km/s from about 5 to 25 kpc (Eilers et al. 2019, as recalled), so the mass enclosed grows roughly ∝ r.

Where does each law hold, and what sets the boundary? Can one small model show both regimes and where one turns into the other?

First estimates

These numbers come from a scoping script and from recall. Check them before they go into a README or a test.

  • Where the Sun's Kepler law ends. It ends at the Sun's tidal (Jacobi) radius in the Galaxy, r_J = [GM☉ / (4A(A − B))]^(1/3). With Oort constants A = 15.3 and B = −11.9 km s⁻¹ kpc⁻¹ (Bovy 2017), this gives 1.37 pc, about 2.8×10⁵ AU, near the Oort cloud's outer edge.
  • The Galaxy's own Kepler region. Inside Sgr A*'s sphere of influence, r_h = GM_BH/σ², which is about 1.8 pc for σ = 100 km/s, stars orbit the way planets do.
    • Take GRAVITY's 2020 elements for S2: a = 125.058 mas, P = 16.0455 yr, e = 0.8846, R₀ = 8246.7 pc.
    • Kepler's third law then returns 4.26×10⁶ M☉. That is their fitted mass, so this is a consistency check, not a measurement.
    • General relativity's precession formula gives 12.1′ per orbit. They detected 12′.
    • So the analogy holds within about 2 pc of the Galaxy's centre, about the same reach as the Sun's own Kepler law, and fails across the disk.
  • The missing mass in one line. Treat the baryons, about 6×10¹⁰ M☉, as a point mass. That gives 177 km/s at R₀ and 102 km/s at 25 kpc, against about 230 measured.
  • How exact Kepler is in the Solar System. From JPL's approximate mean elements (1800–2050), n²a³ matches the IAU GM☉ to about 0.1%.
    • Jupiter's and Saturn's residuals are close to their own masses.
    • Uranus's and Neptune's residuals, about 1×10⁻³, are closer to the mass of the planets inside their orbits.
    • A test should model both effects (and say which frame the elements are in) before it claims a tolerance.
  • An acceleration scale. In galaxies the empirical crossover is an acceleration: g† ≈ 1.2×10⁻¹⁰ m s⁻² in the radial acceleration relation (McGaugh, Lelli and Schombert 2016).
    • The Sun's own pull falls to g† at about 7,000 AU, inside its tidal radius.
    • Wide-binary tests at those separations disagree (Chae 2023 against Banik et al. 2024), so the study should report this as contested and not build on it.

What does repeat

Flattened, rotating disks appear at both scales, because gas loses energy but keeps its angular momentum. The planets' orbits are inclined by a few degrees. The Galaxy's thin disk has a height-to-length ratio of about 0.1. A short paragraph, with sources, is enough.

First steps

  • orbits/model.js, covering:
    • Kepler's third law, with the planet's own mass and the interior-mass corrections;
    • the Jacobi radius;
    • S2's orbit and its precession;
    • a Milky Way mass model with Sgr A* added. For example, Bovy's MWPotential2014 (power-law bulge with a cutoff, Miyamoto–Nagai disk, NFW halo). It should give v(r) from 10⁻³ pc to 100 kpc and the log slope d ln v / d ln r.
  • Mark where that slope is −1/2 and where it is near 0. The boundary should come out as a ratio of enclosed mass to central mass, not a fixed size.
  • Check against:
    • the IAU 2015 nominal GM☉ and JPL's elements;
    • GRAVITY 2020 (S2);
    • Eilers et al. 2019 (the flat curve and its gradient of about −1.7 km s⁻¹ kpc⁻¹);
    • Bovy 2017 (the Oort constants).
  • Say what this does not settle. Dark matter against modified gravity is the Witness Map's and Bound the gravity-only witnesses for dark matter besides black hole horizons #3's ground. This study only measures where the Keplerian nesting stops.

Activity

  1. ahrazzle commented on Oct 19, 2026

    @ahrazzle
    MemberAuthor

    PR #37 has the study. The issue's first estimates mostly held, with these corrections:

    • The planets' total mass is not a bound. It sets the size of the Kepler departures. A ring-averaged estimate puts the outer planets at 1.4 to 1.5×10⁻³, above it.
    • Banik et al. 2024. They prefer Newtonian dynamics at 19σ and exclude MOND at 16σ; these are two different numbers.
    • New in the study: how much of the tidal radius a body keeps.
      • A Hill-problem integrator puts the half-escape radius at about 0.45 r_J for the Sun, which is 1.26×10⁵ au, and 0.53 r_H around a planet.
      • The Galaxy's vertical tide is 7.6 times a point mass's, relative to the orbit's own rate, and loses polar orbits sooner.

    Generated by Claude Code

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions