Holistic view of our universe — Expanding Solar System Resonance Theory

The orbits of the eight planets, the wobble of Earth’s axis, and the rhythm of ice ages are all governed by the same physics. This holistic view puts them in one place: standard Newtonian dynamics carrying every planet from a single cited J2000 state, and Earth’s long periods read as beats against its lunisolar precession clock — one integrated account of solar system dynamics and Earth’s climate record, in a form you can run. The preprint, “The Expanding Solar System Resonance Theory: A Geocentric Model of Earth’s Precession Cycles, Eclipses and Deep-Time Climate”, is on Research Square — read the preprint → . Its thesis in one sentence: one seed, one derived clock and one codebase produce Earth’s precession, timekeeping, eclipses and the orbital climate lines together, and the chain’s deep-time evolution is gated by the rock record.

How the model works, in 60 seconds. Two things carry the model:

• The N-body engine — standard Newtonian gravity with the first-order relativistic correction, integrated from one cited J2000 state with zero fitted constants. It supplies every planet’s orbit, the secular modes, the 405.6-kyr eccentricity metronome, and Earth’s own orbit: its eccentricity, inclination, node and perihelion of date.
• The clock — Earth’s mean lunisolar precession period, 25,771.4 years at J2000, derived from the model’s own year laws (the IAU value to eight parts in a million) and lengthening through geological time as tides slow the spin.

Earth’s other long periods follow as beats of the clock against its orbit’s own motions. A third layer — a fitted timing anchor of 335,317 years, calibrated on the 1246 AD perihelion–solstice alignment — is the unit of the small periodic corrections that bring the time-domain machinery onto the observed cardinal points, day lengths and eclipses. It is bookkeeping, not a cycle: nothing in the sky returns after one anchor interval.

Nothing is tuned per phenomenon. Change the anchor by even a few years and the cardinal points, the day lengths and the eclipses stop matching the record together. That is what makes the model falsifiable. (The clock was ~21,699 yr at 380 Ma; see Expanding Resonance for the deep-time evolution.)

See How It Works for the derivation methodology and Supporting Evidence for what aligns with and where the model differs from current science.

How far it holds — the same machinery, checked against independent records across six orders of magnitude in time:

The numbers were derived from measurement, not theorized — and the same engine produces all of them.

One Seed, One Clock

The model has two engines and one clock. The planets’ orbits — and Earth’s own eccentricity, inclination and perihelion of date — come from a Newtonian N-body integration of one cited J2000 state. Earth’s spin-side periods are read against its mean lunisolar precession period, derived from the model’s own year laws. Two of Earth’s precession motions rotate in opposite directions:

Earth’s two counter-rotating precessions. The perihelion direction advances counter-clockwise around the Sun at T ≈ ~111,635 years (apsidal precession); Earth’s Wobble Center circles Earth clockwise at T ≈ ~25,771 years (axial precession, the clock). Their ratio reads 4.332 today — a reading that wanders, not a constant.

Because the two motions run in opposite directions their rates add, and their beat is the ~20,938-year perihelion-of-date cycle; the obliquity cycle is the clock’s beat against the orbit plane’s nodal turn. Every other quantity the model publishes — the precession of the equinoxes, the obliquity history, the eccentricity cycles, the lengths of days and years, the eclipse geometry, the climate formula’s line positions — is read from the same two engines against the same clock, and the planets’ periods are the engine’s own eigenfrequencies. The simulator draws the two motions as two reference points, Earth’s wobble centre and its perihelion marker; they are the picture, not the physics. A 3D simulation visualises it all in one interactive view.

Why This Matters

Modern celestial mechanics — JPL Horizons, Laskar’s secular theory, high-precision ephemerides — achieves sub-arcsecond accuracy on Myr timescales and remains the right instrument for high-precision position prediction. This framework claims no better accuracy and no new physics. What it adds is integration:

1. Each phenomenon normally has its own model. Precession, obliquity, eccentricity, day and year lengths, eclipse timings and the climate spectrum are all driven by the same gravity, but are conventionally modelled independently, with separate parameterizations that never have to agree with one another.
2. Here they share one causal chain. One cited J2000 state and one derived clock feed every published quantity. Nothing is tuned per phenomenon, so a change anywhere has to stay consistent everywhere — which is what makes the whole thing falsifiable at once rather than one claim at a time.
3. The chain can be perturbed end to end. Multiply Jupiter’s GM by 1.01 and re-integrate: every planet’s apsidal rate moves, Mercury’s by +1.5″/cy — exactly one percent of Jupiter’s first-order Laplace–Lagrange share of Mercury’s precession — and Saturn’s window rate deepens its retrograde phase. The answer is not looked up; it falls out of the same run.
4. No comparable unified 3D view exists. Existing tools handle orbital mechanics or Milankovitch cycles in isolation. None bring axial precession, obliquity oscillation, perihelion precession, eccentricity cycles, day and year length variation, eclipses and all eight planets’ orbits into a single interactive scene.

Observation First — A Bottom-Up Approach

Standard paleoclimate theory takes the climate record as its starting point. The 100-kyr, 41-kyr, and 23-kyr cycles observed in ice cores and ocean sediments are interpreted by selecting which combinations of planetary motions best fit each peak — eigenmode beats from classical Laplace–Lagrange theory. The math is rigorous, but the climate-to-planet attribution is fitted to the data. Several major puzzles remain unexplained: the 100,000-year problem (eccentricity forcing is too weak to drive ice ages directly), the absence of the theoretically dominant 400-kyr cycle in post-MPT climate records, and the Mid-Pleistocene Transition ~1 million years ago.

This holistic view reverses the direction. It starts from directly observed orbital motion — Earth’s measured precession periods, every planet’s measured orbital cycles — anchored against JPL Horizons / WebGeoCalc ephemeris data (1800–2100 AD, every target inside 0.012°) and historical astronomical records. From these observations the model’s own N-body engine yields the planets’ secular modes, and Earth’s spin clock is derived from its year laws. Only then are the beats of those quantities — the climate formula’s 28 orbital lines — overlaid onto the climate record. The line periods are fixed by the model’s dynamics, not fitted to climate — only the per-line amplitudes are fitted to LR04 / CENOGRID / EPICA / CenCO2PIP.

The framework makes a structural commitment the standard approach does not: the orbital layer carries only the engine’s own beat lines, a sharply falsifiable claim. A climate peak with no counterpart among those beats would refute the orbital layer.

The canonical Climate Formula on LR04 δ¹⁸O over the past 700,000 years. R² = 0.779 post-MPT is an in-window description — the 28 engine-derived orbital lines with their amplitudes fitted inside the window; the model’s own orbital histories predict 0.357 of the record out of sample (T7, pre-registered).

From Software to Theory

This framework began as a software project — a 3D simulation that reproduces the positions of the Sun, Moon, and all seven planets, verified against JPL Horizons ephemeris data and historical astronomical observations. The same run simultaneously produces obliquity, perihelion precession, and the planets’ inclinations and eccentricities — quantities normally computed by separate models. Its inputs are kept in three ledgers rather than a single parameter count (Mathematical Foundations): the planetary side has no free parameters at all, and the fitted constants live only in the time-domain correction stack.

From this, a theoretical framework emerged:

• Observable patterns in obliquity, eccentricity, orbital longitudes, and planetary motions could be captured in closed-form formulas
• Precise measurements of day lengths and year lengths in the 3D model also reduced to formulas
• A clock emerged: Earth’s axial precession period, derived from the year laws, against which every other Earth period is a beat — and which lengthens through geological time on the measured tidal history
• The planets ride dynamics: an N-body engine seeded by one cited J2000 state supplies their orbits, their secular modes and their long-period descriptors, with no fitted constant on that path

The simulation also serves as a test bed: existing results such as Souami & Souchay’s (2012) invariable plane definitions can be verified, and three falsifiable deep-time predictions — each naming the observation that would falsify it — can be checked against future observations.

What is at risk

The model stands on three pre-registered legs, each naming what would end it:

• The clock against the rock record. Earth’s precession rate is composed from the spin history and the receding Moon’s torque. It meets the cyclostratigraphic inference of Wu et al. 2024 at 650 Ma to 0.2 %, and the Precambrian precession constants at 1.4 and 2.46 Ga are carried as gates, where pure spin scaling already fails. The formula’s precession-band and obliquity lines ride that lengthening clock; its eccentricity-band lines are planetary beats and do not — every newly dated Precambrian section tests the split (Expanding Resonance).
• Historical-era exactness. One ΔT formula — pure-tidal Moon recession plus a GIA channel anchored on satellite gravimetry, with zero parameters fitted to eclipse data — does three things at once. It places the umbra within the ±4-hour scan window for 21/26 documented solar eclipses (−762 to 2026 CE), gives a 20.2-min mean |residual| on the 267-event Stephenson 2016 lunar set, and rejects the full Munk–MacDonald non-tidal speedup (Timekeeping).
• The two expansions’ consistency. Under the measured solar-mass history the 405-kyr cycle in dated Precambrian strata weighs the ancient Sun: μ(2.48 Ga) = 1.00 ± 0.07 (Expanding Resonance).

Two retrodictions come with no deep-time constraint in the fit: the Earth–Moon genesis at the rigid Roche limit at the giant-impact epoch (~4.498 Ga) and the Wells 1963 Devonian day count to 0.01 %. The Sun and the Moon behind the eclipse tests are themselves assembled from the framework’s structure — The Derived Sun with zero fitted solar constants, The Derived Moon with every Meeus constant derived, attributed or anchored by design.

On the climate record the model commits to a finite line list — 28 beats whose periods are its own modes and clock, only the amplitudes fitted per regime. Every significant LR04 peak has a counterpart among them; the in-window fits (R² = 0.779 on post-MPT LR04, 0.730 on EPICA CO₂, 0.692 on CenCO2PIP 0–66 Ma) are descriptions, and the model’s own orbital histories predict 0.357 of post-MPT LR04 out of sample (T7, pre-registered). Classical insolation features alone explain R² = 0.2362 of the same record, 3× less than the lines, and add no cross-window-stable gain on top of them; the 100-kyr band reads as the planets’ eccentricity beats, not direct Earth-eccentricity forcing (Climate Formula).

An open framework, not a closed theory

The model is offered as a framework for testing scientific theories, not a final word. Every observable in the published literature is reproduced; the two departures from current consensus (the origin of the 100,000-yr glacial cycle, and the Munk-MacDonald non-tidal speedup in pre-1900 length of day) and the one pre-registered deep-time prediction (the two-tier scaling of the climate lines) are catalogued with their discriminating tests.

All data, formulas, and the 3D simulation are publicly available. We invite independent replication and contradicting evidence — the framework is most useful when others apply it to their own data and report what they find.

Read more on Supporting Evidence → An open framework — invitation to test

Explore the Model

Start reading: How It Works →

Quick Facts

• The engine: Newtonian N-body from one cited J2000 state, zero fitted constants — it carries every planet’s orbit and Earth’s own eccentricity, inclination, node and perihelion of date (Moon & Planets)
• The clock — axial precession: 25,771.4 years at J2000 (derived; lengthens through geological time)
• Apsidal precession: ~111,635 years (4.332 clock periods — a reading that wanders, between 1.08 and 9.81)
• Perihelion precession (of date): ~20,938 years (0.8124 clock periods)
• Obliquity cycle: ~41,224 years (the beat of the clock against the orbit plane’s nodal turn)
• Obliquity range: ~22.21° – ~24.72°
• The fitted anchor: 335,317 years — the unit of the correction bases, 13.011 clock periods; bookkeeping, not a cycle