Lecture 04: N-Body Systems, Invariants, and Close Encounters — Lecture Notes

Terms Developed in Context

The terms below connect the mathematical model, implementation choices, computed diagnostics, and limits of the scientific claim.

N-body system

A set of mutually gravitating particles whose accelerations depend on every other particle. In this lecture it is used to address How can conservation diagnostics reveal a timestep that looks visually plausible?

Model connection: Read the term against Invariant and the governing relation \(\ddot{\mathbf r}_i=G\sum_{j\ne i}m_j\frac{\mathbf r_j-\mathbf r_i}{|\mathbf r_j-\mathbf r_i|^3}\). The relevant computational evidence is a controlled output, residual, invariant, or refinement sequence rather than visual smoothness alone.

Interpretive limit: Good invariant conservation is necessary but not sufficient: conserved errors can coexist with wrong initial conditions, wrong frames, or missing forces. This distinction separates numerical behavior from physical accuracy and model adequacy.

Invariant

A quantity such as total energy or angular momentum that the continuous model conserves. In this lecture it is used to address How can conservation diagnostics reveal a timestep that looks visually plausible?

Model connection: Read the term against Close encounter and the governing relation \(\ddot{\mathbf r}_i=G\sum_{j\ne i}m_j\frac{\mathbf r_j-\mathbf r_i}{|\mathbf r_j-\mathbf r_i|^3}\). The relevant computational evidence is a controlled output, residual, invariant, or refinement sequence rather than visual smoothness alone.

Interpretive limit: Good invariant conservation is necessary but not sufficient: conserved errors can coexist with wrong initial conditions, wrong frames, or missing forces. This distinction separates numerical behavior from physical accuracy and model adequacy.

Close encounter

A near approach that creates a short dynamical timescale and strong acceleration. In this lecture it is used to address How can conservation diagnostics reveal a timestep that looks visually plausible?

Model connection: Read the term against Center-of-mass frame and the governing relation \(\ddot{\mathbf r}_i=G\sum_{j\ne i}m_j\frac{\mathbf r_j-\mathbf r_i}{|\mathbf r_j-\mathbf r_i|^3}\). The relevant computational evidence is a controlled output, residual, invariant, or refinement sequence rather than visual smoothness alone.

Interpretive limit: Good invariant conservation is necessary but not sufficient: conserved errors can coexist with wrong initial conditions, wrong frames, or missing forces. This distinction separates numerical behavior from physical accuracy and model adequacy.

Center-of-mass frame

Coordinates measured relative to the system barycenter and its bulk motion. In this lecture it is used to address How can conservation diagnostics reveal a timestep that looks visually plausible?

Model connection: Read the term against Chaos and the governing relation \(\ddot{\mathbf r}_i=G\sum_{j\ne i}m_j\frac{\mathbf r_j-\mathbf r_i}{|\mathbf r_j-\mathbf r_i|^3}\). The relevant computational evidence is a controlled output, residual, invariant, or refinement sequence rather than visual smoothness alone.

Interpretive limit: Good invariant conservation is necessary but not sufficient: conserved errors can coexist with wrong initial conditions, wrong frames, or missing forces. This distinction separates numerical behavior from physical accuracy and model adequacy.

Chaos

Exponential sensitivity of trajectories to small differences in initial conditions. In this lecture it is used to address How can conservation diagnostics reveal a timestep that looks visually plausible?

Model connection: Read the term against N-body system and the governing relation \(\ddot{\mathbf r}_i=G\sum_{j\ne i}m_j\frac{\mathbf r_j-\mathbf r_i}{|\mathbf r_j-\mathbf r_i|^3}\). The relevant computational evidence is a controlled output, residual, invariant, or refinement sequence rather than visual smoothness alone.

Interpretive limit: Good invariant conservation is necessary but not sufficient: conserved errors can coexist with wrong initial conditions, wrong frames, or missing forces. This distinction separates numerical behavior from physical accuracy and model adequacy.

Problem and physical meaning

Mutual gravity couples all particles, and close approaches shorten local dynamical timescales. WHFast is effective for hierarchical systems with a dominant primary; adaptive IAS15 or hybrid methods are more suitable when close encounters violate that structure.

How can conservation diagnostics reveal a timestep that looks visually plausible? Begin from the scientific observable or state variable, and name what is directly specified versus inferred. The model should be simple enough to validate, but not quietly presented as the entire astrophysical system.

Model and derivation

The governing expression for this lecture is:

Define the symbols and the independent variable before numerical evaluation. Dimensional analysis checks that both sides have matching units; if the model is dimensionless, state the adopted scale factors and the transformation back to physical units.

Mutual gravity couples all particles, and close approaches shorten local dynamical timescales. WHFast is effective for hierarchical systems with a dominant primary; adaptive IAS15 or hybrid methods are more suitable when close encounters violate that structure.

Numerical implementation

Use the Sun-Earth-Jupiter benchmark with measured mass parameters and declared idealized circular geometry; compare fixed steps over 12 years; monitor energy, angular momentum, and center-of-mass momentum; rerun a close-encounter case with an appropriate hybrid method.

Keep physical quantities and numerical arrays conceptually separate. Convert units at well-defined boundaries, preserve raw inputs, and record dtype, solver, tolerances/timestep, grid, and software versions. A numerical method is part of the model evidence, not an invisible implementation detail.

Worked, programmatically computed example

Across 12 years, the three-body WHFast relative energy error falls from 8.56e-12 at dt=5.707 d to 5.33e-13 at dt=1.427 d; angular-momentum and scaled momentum errors are tabulated separately.

These values are generated in src/astr474_computations.py, then shared with lecture/lab/assignment renderers. Re-run the generator after changing an input; do not hand-edit a displayed result without recomputing dependent artifacts.

At least one independent check is required: an analytic solution, a second numerical method, a conservation law, or a controlled convergence sequence. When these disagree, diagnose units, frame, discretization, source terms, and model scope before selecting a result.

Error budget and limitations

Good invariant conservation is necessary but not sufficient: conserved errors can coexist with wrong initial conditions, wrong frames, or missing forces.

Classify each uncertainty before combining terms: local/global solver error, timestep or spatial discretization, Monte Carlo sampling, parameter uncertainty, measurement uncertainty, and model discrepancy are not interchangeable. Conservation can test internal consistency but does not prove the model is physically complete.

Connection to evidence and research practice

Lab 03 varies timestep in REBOUND and asks students to defend the integrator choice for an encounter.

For real parameters, the source record distinguishes the fetched product and epoch from derived values. For synthetic experiments, the initial conditions, generator, seed, and exact file are recorded. A simulation is not an observation, even when its parameters resemble a real object.

Misconceptions to avoid

• More displayed digits do not imply more physical accuracy.
• A successful solver exit or smooth plot does not establish convergence.
• Agreement between two solvers with the same wrong model is not independent physical validation.
• A formal covariance or Monte Carlo interval is conditional on the adopted model and noise process.

Study and review questions

1. How can conservation diagnostics reveal a timestep that looks visually plausible?
2. Which numerical parameter would you vary first, what observable would you track, and how would you know the sequence reached an asymptotic regime?
3. Design one independent check and state what failure would change your scientific interpretation.

Preparation for the next lecture: explain how this method's assumptions change when the problem becomes more coupled, stochastic, or data-driven.

Lecture-specific derivation and assumptions

For point masses, Newton's law gives a_i=G*sum_{j!=i} m_j (r_j-r_i)/|r_j-r_i|^3. Pairwise forces cancel in the total momentum derivative, and their central direction makes total angular momentum constant; a time-independent Newtonian potential conserves total energy. WHFast uses a split Hamiltonian: an exactly solvable dominant Kepler part and perturbing interaction kicks. A close encounter weakens that hierarchy and introduces a short encounter timescale, so fixed steps must resolve it or a different integrator is needed.

Algorithm trace and computed-output interpretation

Run the declared Sun-Earth-Jupiter initial state at 64, 128, and 256 steps per Earth year for the same 12-year span. Tabulate relative energy and angular-momentum errors plus momentum error scaled by the sum of particle momentum magnitudes. Compare adjacent ratios and timestep, then repeat an encounter case with IAS15. Do not compare raw momentum residuals without the scale because the center-of-mass frame can make net momentum small by construction.

Computed result: Across 12 years, the three-body WHFast relative energy error falls from 8.56e-12 at dt=5.707 d to 5.33e-13 at dt=1.427 d; angular-momentum and scaled momentum errors are tabulated separately.

A bounded oscillatory energy error can be consistent with symplectic integration even when the instantaneous error is not monotone. Conservation checks detect some implementation defects but cannot detect a consistently wrong frame, mass, or force model. The three-body geometry here is synthetic and intentionally idealized.

Two-meeting teaching arc

Each weekly module is taught in two 75-minute meetings. The sequence below includes a formative check in each meeting and is intended to use the lecture's worked computation, not merely present it.

1. Meeting 1 (75 min): 10 min N-body motivation; 20 min acceleration and invariant derivations; 20 min Hamiltonian splitting; 20 min choose diagnostics for a hierarchical system; 5 min encounter-scale estimate.
2. Meeting 2 (75 min): 10 min recap; 20 min close-encounter limitations; 25 min inspect timestep/conservation outputs; 15 min compare IAS15 and WHFast use cases; 5 min state a validation criterion.

Targeted study questions

1. Why can a symplectic method show bounded energy oscillation rather than monotonic drift?
2. Estimate an encounter timescale from impact parameter b and relative speed v; what step fraction would you test?
3. Which two diagnostics can remain good while the predicted orbit is still scientifically wrong?