Conley-Zehnder Indices of Spatial Rotating Kepler Problem

This paper provides a complete symplectic-topological classification of periodic orbits in the spatial rotating Kepler problem by introducing a new coordinate system based on the Laplace-Runge-Lenz vector, computing their Conley-Zehnder and Robbin-Salamon indices, and determining their contributions to symplectic homology via the Morse-Bott spectral sequence.

The Gist

Imagine you are watching a cosmic dance. In the center of the stage sits a massive, stationary star (like our Sun). Orbiting it is a tiny, weightless dancer (like a comet or a satellite). This is the classic Kepler Problem: a simple, elegant dance governed by gravity.

But now, imagine the entire stage is spinning. The star isn't just sitting there; the whole universe is rotating around a vertical axis. This is the Rotating Kepler Problem. The dancer is no longer just orbiting a static point; they are fighting against the spin of the stage itself.

This paper by Dongho Lee is like a master choreographer's notebook. It doesn't just watch the dance; it tries to classify every possible move, predict how the dancer will spin, and assign a unique "score" to every routine to understand the deep mathematical structure of the universe.

Here is a breakdown of the paper's main ideas using simple analogies:

1. The Map of the Dance Floor (Classification)

In the old days, describing these orbits was like trying to describe a complex dance using only words like "spin left" or "jump high." It was messy.

Lee introduces a new GPS system for these orbits. He uses two special tools:

• Angular Momentum: How much the dancer is spinning around the center.
• The Laplace-Runge-Lenz Vector: A fancy way of saying "the shape and orientation of the orbit." Think of it as a compass that always points to the closest point of the orbit (the perigee).

The Big Discovery: Lee proves that if you know these two numbers, you can pinpoint the orbit on a giant, invisible map. It turns out that all possible orbits fit perfectly onto a shape that looks like two spheres glued together ($S^2 \times S^2$). It's like realizing that every possible dance move in the universe corresponds to a specific coordinate on a globe.

2. The Three Types of Moves

Once the map is drawn, Lee identifies three main types of dancers that keep coming back to the same spot (periodic orbits):

• The Retrograde Dancer ($\gamma_+$): This dancer spins against the rotation of the stage. They are fast, close to the center, and move counter-clockwise while the stage spins clockwise. They are the "champions" of the dance floor.
• The Direct Dancer ($\gamma_-$): This dancer spins with the rotation of the stage. They are slower, further out, and move clockwise.
• The Vertical Diver ($\gamma_c$): This is a unique move found only in the 3D (spatial) version. Imagine the dancer falling straight down through the center of the stage, hitting the "floor," and bouncing straight back up. It's a collision orbit, but in this math world, we can smooth out the crash so the dance continues forever.

3. The "Score" of the Dance (Conley-Zehnder Indices)

This is the most technical part of the paper, but think of it as assigning a difficulty score to every dance routine.

In symplectic topology (a branch of math that studies shapes and flows), we need to know how "twisted" an orbit is. The Conley-Zehnder index is that score.

• If the dancer does a simple loop, the score is low.
• If they do a complex series of twists and turns, the score goes up.

Lee calculates these scores for the Retrograde, Direct, and Vertical dancers.

• The Twist: He finds that the "spatial" (3D) dancers have scores that are exactly double what they would be if the dance were flat (2D). It's like adding a third dimension to a gymnastics routine doubles the complexity of the twists.
• The Vertical Diver: Surprisingly, the score for the vertical collision orbit is always a clean multiple of 4 ($4N$). It's a very stable, predictable move.

4. The "Family Reunion" (Morse-Bott Families)

Sometimes, the dancers don't just do one specific move; they do a whole family of moves that look slightly different but are essentially the same.

Imagine a group of dancers all spinning at the same speed but starting at slightly different angles. They form a "family" (called a $\Sigma_{k,l}$ family).

• In the past, math tools struggled to count these families because they were "degenerate" (too similar to each other).
• Lee introduces a new coordinate system based on that "compass" vector (Laplace-Runge-Lenz) to untangle this mess. He proves these families are stable and calculates their score (Robbin-Salamon index) as $4k - 1/2$.

5. Why Does This Matter? (Symplectic Homology)

You might ask, "Why do we care about these scores?"

Think of Symplectic Homology as a way to count the "holes" or "tunnels" in the shape of the universe. The dancers (orbits) are the generators of this count.

• Lee shows that the dancers he found (Retrograde, Direct, Vertical) perfectly match the "holes" predicted by the theory.
• It's like finding the exact number of keys needed to open every lock in a giant castle. By calculating the indices, Lee proves that the Rotating Kepler Problem is a perfect, geometric realization of these abstract mathematical concepts.

The Takeaway

This paper takes a problem that has puzzled mathematicians for centuries (how things move under gravity in a spinning universe) and solves it by:

1. Mapping every possible orbit onto a clear geometric shape.
2. Scoring every orbit to see how complex it is.
3. Proving that these orbits are the building blocks of a deeper mathematical structure (Symplectic Homology).

It's a bridge between the physical world of planets and satellites and the abstract world of pure geometry, showing that even in a chaotic, spinning universe, there is a perfect, orderly rhythm waiting to be discovered.

Technical Summary

Here is a detailed technical summary of the paper "Conley-Zehnder Indices of Spatial Rotating Kepler Problem" by Dongho Lee.

1. Problem Statement

The paper investigates the spatial rotating Kepler problem, a dynamical system describing the motion of a massless body under the gravitational influence of a fixed central body, observed from a rotating reference frame. This system serves as a limiting case of the circular restricted three-body problem (where the secondary body has negligible mass).

The primary objectives are:

1. To provide a complete classification of periodic orbits in the spatial setting (3D), extending previous results known for the planar (2D) case.
2. To compute the Conley-Zehnder (CZ) indices for non-degenerate periodic orbits and the Robbin-Salamon (RS) indices for degenerate families of orbits.
3. To establish the connection between these indices and the symplectic homology of the cotangent bundle $T^*S^3$, specifically identifying the orbits as generators of the homology groups.

A significant challenge addressed is the coordinate degeneracy that arises when moving from the planar to the spatial setting, particularly regarding the standard Delaunay coordinates used in previous literature.

2. Methodology

A. Symplectic Setup and Regularization

• Hamiltonian Formulation: The system is governed by the Jacobi energy $H = E + L_3$, where $E$ is the Kepler energy and $L_3$ is the angular momentum component along the rotation axis.
• Moser Regularization: To handle the singularity at the origin (collision orbits) and compactify the energy hypersurfaces, the author employs Moser's regularization. This maps the energy level set of the Kepler problem to the unit cotangent bundle of the 3-sphere, $ST^*S^3$, effectively transforming the problem into a geodesic flow on $S^3$.
• Moduli Space Parametrization: The author utilizes the angular momentum vector ($L$) and the Laplace-Runge-Lenz (LRL) vector ($A$) to parametrize the moduli space of Kepler orbits. A key bijection $\Phi: M_E \to S^2 \times S^2$ is established, mapping an orbit to the pair $(\sqrt{-2E}L - A, \sqrt{-2E}L + A)$.

B. Index Computation Techniques

• Conley-Zehnder Index ($\mu_{CZ}$): Calculated for non-degenerate orbits by analyzing the linearized Hamiltonian flow. The author distinguishes between the full linearized flow and the transverse flow (restricted to the contact structure), which is the standard definition for Reeb orbits in contact manifolds.
• Robbin-Salamon Index ($\mu_{RS}$): Used for degenerate orbits (families of periodic orbits). The calculation involves the crossing form of the path of symplectic matrices associated with the linearized flow.
• Morse-Bott Spectral Sequence: For degenerate families (where orbits form manifolds rather than isolated points), the author uses the Morse-Bott spectral sequence to relate the local Floer homology of the critical manifolds to the global symplectic homology.
• Coordinate Systems:
  • Cylindrical Coordinates: Used for planar circular orbits.
  • Delaunay Coordinates: Used for generic non-planar orbits but found to be partially degenerate in the spatial case.
  • LRL-based Coordinates: A novel coordinate system introduced by the author based on the Laplace-Runge-Lenz vector ($A_3$) to resolve the degeneracies of Delaunay coordinates in the spatial setting, allowing for a rigorous proof of the Morse-Bott property.

3. Key Contributions and Results

A. Classification of Periodic Orbits

The paper classifies periodic orbits for a fixed Jacobi energy $c < -3/2$ into three categories:

1. Planar Circular Orbits:
   • Retrograde ($\gamma_+$): Rotates opposite to the frame rotation.
   • Direct ($\gamma_-$): Rotates with the frame rotation.
   • These are isolated non-degenerate orbits.
2. Vertical Collision Orbits ($\gamma_{c\pm}$): Orbits oscillating along the rotation axis ($q_3$-axis) passing through the origin. These are unique to the spatial problem and are non-degenerate.
3. Morse-Bott Families ($\Sigma_{k,l}$): Resonant orbits where the Kepler period $\tau$ and the rotation period satisfy a resonance condition $k\tau = 2\pi l$. These form families diffeomorphic to $S^3 \times S^1$.

B. Computation of Indices

The author derives explicit formulas for the indices of the $N$-th iterates of these orbits:

1. Planar Circular Orbits ($\gamma_\pm$):
   $$\mu_{CZ}(\gamma_N^\pm) = 2 + 4 \left\lfloor \frac{N (-2E)^{3/2}}{(-2E)^{3/2} \pm 1} \right\rfloor$$

   • Observation: The indices are exactly twice those of the corresponding planar orbits, reflecting the additional spatial dimension.

2. Vertical Collision Orbits ($\gamma_{c\pm}$):
   $$\mu_{CZ}(\gamma_N^{c\pm}) = 4N$$

   • Observation: The index is independent of the energy $E$ and scales linearly with the iteration number $N$.

3. Morse-Bott Families ($\Sigma_{k,l}$):
   The Robbin-Salamon index for the family $\Sigma_{k,l}$ is:
   $$\mu_{RS}(\Sigma_{k,l}) = 4k - \frac{1}{2}$$

   • Significance: This index is crucial for determining the degree shift in the spectral sequence.

C. Morse-Bott Property

The paper proves that the families $\Sigma_{k,l}$ satisfy the Morse-Bott condition (the critical set is a submanifold, and the Hessian is non-degenerate in the normal direction).

• Novelty: The author demonstrates that standard Delaunay coordinates fail to cover planar orbits in the spatial case due to degeneracy. By introducing the LRL-based coordinate system, the author successfully proves the Morse-Bott property for all relevant orbits, including planar ones.

D. Relation to Symplectic Homology

The computed indices are shown to align perfectly with the grading of the $S^1$-equivariant symplectic homology of $T^*S^3$.

• The generators of the homology groups correspond to the periodic orbits found:
  • Degree 2: Retrograde orbit ($\gamma_+$).
  • Degree $4k$: Vertical collision orbits ($\gamma_{c\pm}^k$).
  • Degree $4k+2$: Combinations of direct and retrograde iterates.
• This confirms that the spatial rotating Kepler problem provides a geometric realization of the generators of the symplectic homology of $T^*S^3$.

4. Significance

1. Extension to 3D: This work successfully extends the symplectic-topological understanding of the rotating Kepler problem from the well-studied planar case to the full spatial (3D) case, revealing new orbit types (vertical collisions) and index behaviors.
2. Resolution of Degeneracy: The introduction of the LRL-based coordinate system solves a technical bottleneck in applying action-angle variables to spatial celestial mechanics problems, offering a tool applicable to other central force systems.
3. Symplectic Homology Verification: By explicitly computing the indices and matching them to the known homology of $T^*S^3$, the paper provides a concrete, dynamical verification of abstract symplectic invariants. It bridges classical mechanics (Kepler problem) with modern Floer theory.
4. Foundation for Further Study: The complete classification and index computation provide the necessary groundwork for studying bifurcations, stability, and the existence of periodic orbits in more complex systems, such as the full restricted three-body problem.

In summary, Dongho Lee's paper provides a rigorous symplectic-topological profile of the spatial rotating Kepler problem, utilizing advanced regularization and coordinate techniques to classify orbits and compute their indices, thereby linking classical dynamical systems directly to the generators of symplectic homology.