R&D Capabilities — Generative Physics

Physics Model Reference

Every object Cosmos Genesis generates — from individual stars to galaxies of trillions — is produced and evolved by a stack of peer-reviewed physics models. This page covers the generative engine: stellar evolution, galactic dynamics, binary systems, planetary formation, and compact objects.

For applied capabilities built on top of this engine — space weather, radiation environments, satellite tracking, and mission planning — see the R&D Capabilities section.

Tier 1 — Public capability documentation. This page describes what Cosmos Genesis models and the published science behind it. Algorithm implementations, calibration constants, and pipeline internals are maintained as Tier 3 internal documentation.

How aging works today. Every model on this page runs once, during generation, bringing each object forward to a single requested epoch — it does not keep evolving after that. Subscribing to have a licensed galaxy or system aged forward on a recurring schedule, and keeping systems your users are actively visiting synchronized with the wider simulation, are planned capabilities covered in Deep Time and gated by subscription tier.

Galactic Scale

Universe Generation at Scale

The generation pipeline assembles physically accurate stellar populations for any galaxy morphology — dwarf spheroidals, lenticulars, spiral arms with active nuclei, irregular starbursts. Every star is drawn from a stellar initial mass function (IMF) validated against observed galactic demographics, spanning the full mass range from brown dwarf candidates through O-class supergiants.

Each star is assigned an age, metallicity, and evolutionary stage consistent with its host galaxy's star formation history. Chemical enrichment accumulates across generations — older stellar populations carry lower metallicity, younger ones reflect the iron and alpha-element abundance built up by preceding supernova cycles. Galaxy-scale kinematics follow observed rotation curves, including the flat outer-disk profile that implicates dark matter halos.

At full scale the pipeline produces and individually tracks objects across galaxies of trillions of stars, using distributed compute on EMR Serverless with Iceberg tables on S3 for columnar access at any resolution.

Outputs & Scale

Mass functionKroupa (2001) IMF — validated against galactic stellar census
Metallicity modelAge-metallicity relation with supernova enrichment
Kinematic modelFlat rotation curve — consistent with dark matter halo models
Pipeline scaleIndividual-object tracking to trillions of stars
Output formatIceberg on S3 — queryable via Athena

Published References

  • →Kroupa (2001) — Variation of the initial mass function — MNRAS 322:231

Stellar Evolution

Stellar Life Cycles

Every star in a generated system evolves through its full life cycle within the simulation. Main sequence evolution tracks how luminosity, radius, and surface temperature change over gigayear timescales as hydrogen is consumed in the stellar core.

Post-main-sequence phases — the subgiant branch, red giant branch, and asymptotic giant branch — are each modeled as distinct evolutionary stages with physics-appropriate luminosity amplification and mass loss. Stellar winds carry mass away continuously throughout the giant phases, following the empirical mass-loss scaling established by Reimers for evolved stars. The final evolutionary outcome — white dwarf, neutron star, or stellar-mass black hole — is determined by the star's initial mass.

Every star is evolved forward, once, from its formation age to a chosen simulation epoch — so a generated galaxy presents a realistic distribution of evolutionary stages rather than a frozen snapshot of zero-age main sequence objects. That evolution stops at the requested epoch until the system is aged again.

Outputs & Scale

Timescales covered1 Myr → 13 Gyr (full galactic history)
Mass-loss modelReimers (1975) empirical law
Post-MS luminosity amplificationStage-appropriate factors (SGB → RGB → AGB)
Remnant classificationWhite dwarf · neutron star · black hole by initial mass
Output fieldsMass, luminosity, radius, temperature, evolutionary stage

Published References

  • →Reimers (1975) — Empirical stellar mass-loss formula — A&A Suppl. 22

Stellar Death

Supernovae & Transient Events

High-mass stars end their lives in core-collapse supernovae — explosive events that release approximately 1051 erg of energy into the surrounding interstellar medium. The simulation tracks the expanding blast wave, calculates the affected volume, and assigns damage severity to neighboring systems within the blast radius.

Environmental impact cascades: atmospheric stripping removes significant fractions of planetary envelopes in close-proximity systems, and orbital perturbations from the gravitational kick can destabilize planetary architectures. The severity model scales with distance from the progenitor and the progenitor's mass.

Intermediate-mass stars eject their outer envelopes as planetary nebulae, producing UV-ionized shells described by Strömgren sphere geometry. These ionized regions can disrupt protoplanetary disks and alter habitability conditions in surrounding systems, coupling stellar death to planetary environment physics across light-year scales.

In a generated galaxy, these outcomes are assigned as part of the generation pass — a star already past its death is generated in its post-event state. Authoring a death live, on a star that hasn't reached it yet, is a separate capability under active development; see Stellar Evolution Studio.

Outputs & Scale

Supernova energy~10⁵¹ erg (1 foe) — standard core-collapse yield
Blast radiusScales with progenitor mass — up to ~50 LY for massive progenitors
Damage classificationSevere · moderate · minor by distance fraction
Atmosphere lossComputed for planets within blast radius
Nebula ionizationStrömgren sphere geometry — disrupts nearby disks

Published References

  • →Bethe (1990) — Supernova mechanisms — Reviews of Modern Physics 62:801
  • →Strömgren (1939) — Ionization of interstellar hydrogen — Astrophysical Journal 89:526

Multi-Star Systems

Binary & Multi-Star Dynamics

Over half of all stars in the Milky Way exist in gravitationally bound pairs or higher-order systems. The simulation models the key physical processes that govern how these systems evolve across cosmic timescales.

Tidal forces progressively circularize initially eccentric orbits — the timescale for this process is a steeply nonlinear function of orbital separation, so compact binaries circularize on million-year timescales while wide binaries may retain significant eccentricity through their entire lifetimes. In close binaries, one star can overflow its gravitational sphere of influence — the Roche lobe — transferring mass to its companion at rates that fundamentally alter both stars' evolutionary trajectories.

For the tightest binaries, energy loss through gravitational wave emission steadily shrinks the orbit on gigayear timescales, with merger the eventual outcome for some systems. Planetary orbits in multi-star environments are classified as dynamically stable or unstable using the Holman & Wiegert analytical stability limits, distinguishing S-type (around one star) and P-type (around both) configurations.

Outputs & Scale

Roche lobe geometryEggleton (1983) analytical formula
Tidal circularizationTimescale ∝ (a/R)⁸ — separation-sensitive
Gravitational wave decayOrbital shrinkage tracked to merger
Planetary orbit classificationS-type / P-type / unstable — Holman & Wiegert (1999)
Mass transfer ratesRoche lobe overflow — 10⁻⁸ to 10⁻⁶ M☉/yr range

Published References

Disk Physics

Planetary System Formation

Stellar birth is accompanied by a rotating disk of gas and dust — the protoplanetary disk — from which planets form. The simulation tracks disk evolution through two competing processes: viscous accretion, which drains mass inward toward the star over millions of years following the Shakura-Sunyaev alpha-disk model; and photoevaporation driven by stellar ultraviolet and X-ray flux, which strips mass from the disk surface at a rate that accelerates as the disk thins.

Planet formation depletes the available solid and gas reservoir within the first few million years of a system's life. By the time a typical disk has dispersed — generally within ten million years — the planetary inventory is set. Habitability zone boundaries are computed from stellar luminosity and shift as the star ages and brightens.

Outputs & Scale

Disk accretion modelShakura-Sunyaev α-disk (1973)
Photoevaporation onsetUV/X-ray driven — after ~3 Myr
Typical disk lifetimeComplete dispersal by ~10 Myr
Output fieldsDisk mass, accretion rate, HZ inner & outer boundary
Habitability zone√L scaling from stellar luminosity — updates as star evolves

Published References

Planetary Physics

Planetary Evolution

Planets are not static — their orbits, atmospheres, and surfaces evolve continuously under gravitational, radiative, and thermal forcing. Hot Jupiters in close orbits lose orbital energy to tidal dissipation, drifting gradually inward over hundreds of millions of years. Low-mass planets near luminous stars lose atmospheric gas through XUV-driven photoevaporation, with small worlds potentially shedding entire envelopes on gigayear timescales.

Natural satellites experience outward tidal recession from their parent planets — the Moon's observed recession from Earth (currently ~3.8 cm/yr) is the calibration case for the model. Surface temperatures couple to stellar luminosity evolution, tracking the shift in habitable conditions as a star ages off the main sequence and the habitable zone migrates outward.

Outputs & Scale

Tidal decay modelHot Jupiter orbital decay for a < 0.1 AU
Atmospheric escapeXUV photoevaporation — significant for M < 5 M⊕
Lunar recessionTidal dissipation model — Earth-Moon calibrated
Surface temperatureCoupled to stellar luminosity via Stefan-Boltzmann
Output fieldsSemi-major axis, eccentricity, atmospheric pressure, surface temp

Published References

Orbital Mechanics

N-Body Integration & Ephemerides

Complex gravitational environments — multi-star systems, dense clusters, planet-moon-ring hierarchies, and spacecraft trajectories — require direct numerical integration of the equations of motion for each interacting body. The N-body engine computes gravitational forces between all relevant bodies and integrates orbits forward while conserving energy over long time spans, enabling multi-body simulations of realistically complex systems at scale.

For solar system bodies and spacecraft, SPICE ephemeris data provides verified positional ground truth at epoch — anchoring numerical trajectories to observed positions published by JPL Horizons and the NAIF toolkit.

Outputs & Scale

Energy conservationConserved over full simulation duration — no long-term drift
Ground truth anchorNAIF SPICE kernels — JPL DE440 ephemeris
Positional verificationVerifiable against JPL Horizons at epoch
Use casesMulti-star systems · spacecraft trajectories · moon hierarchies

Published References

  • →Acton (1996) — NAIF SPICE toolkit — Planetary and Space Science 44:65
  • →Park et al. (2021) — DE440 planetary ephemeris — Astronomical Journal 161:105

Exotic Objects

Compact Objects & Black Hole Physics

Black holes formed by stellar collapse grow through accretion of surrounding matter. The maximum rate at which a black hole can accrete is limited by radiation pressure — the Eddington limit — which sets the characteristic timescale for black hole mass doubling and governs how quickly supermassive black holes can reach their observed masses.

Accreting black holes with strong magnetic fields can launch relativistic jets perpendicular to the accretion disk via the Blandford-Znajek mechanism, producing some of the most energetic continuous emission in the universe. Jet power scales with black hole spin and mass, enabling physically grounded modeling of active galactic nuclei at galactic center positions.

For primordial black holes lighter than roughly 1012 kg, Hawking radiation — thermal emission arising from quantum field effects near the event horizon — is the dominant mass-loss mechanism, with evaporation timescales that can be shorter than the age of the universe for the lightest objects.

Outputs & Scale

Accretion modelEddington-limited — characteristic doubling timescale ~45 Myr
Jet mechanismBlandford-Znajek — power scales with spin² and mass²
Spin evolutionPrograde accretion spins up; jets spin down
Hawking evaporationActive for primordial BHs (M < ~10¹² kg)
Output fieldsMass, spin parameter, accretion rate, jet power (erg/s)

Published References

Full Reference List

All physics models are derived from or validated against published peer-reviewed literature. Reviewers can verify each methodology independently from the citations below.

← Back to docs