Linking long-term planetary $N$-body simulations with periodic orbits: application to white dwarf pollution

TL;DR

This study links periodic orbits with long-term N-body simulations to understand debris disc evolution and white dwarf pollution, revealing that certain simplified models do not produce asteroid-star collisions over Gyr timescales.

Contribution

It introduces a comprehensive method connecting periodic orbits with N-body simulations in the CRTBP to analyze long-term asteroid dynamics near the 2:1 resonance.

Findings

No asteroid-star collisions occurred over 14 Gyr in the CRTBP 2:1 model.

Asteroids surviving beyond 35 Myr had pericentres varying by at most 60%.

Results suggest the need to explore more complex orbital architectures for WD pollution.

Abstract

Mounting discoveries of debris discs orbiting newly-formed stars and white dwarfs (WDs) showcase the importance of modeling the long-term evolution of small bodies in exosystems. WD debris discs are in particular thought to form from very long-term (0.1-5.0 Gyr) instability between planets and asteroids. However, the time-consuming nature of $N$-body integrators which accurately simulate motion over Gyrs necessitates a judicious choice of initial conditions. The analytical tools known as \textit{periodic orbits} can circumvent the guesswork. Here, we begin a comprehensive analysis directly linking periodic orbits with $N$-body integration outcomes with an extensive exploration of the planar circular restricted three-body problem (CRTBP) with an outer planet and inner asteroid near or inside of the $2$:$1$ mean motion resonance. We run nearly 1000 focused simulations for the entire age of the Universe (14 Gyr) with initial conditions mapped to the phase space locations surrounding the unstable and stable periodic orbits for that commensurability. In none of our simulations did the planar CRTBP architecture yield a long-timescale ($\gtrsim 0.25$% of the age of the Universe) asteroid-star collision. The pericentre distance of asteroids which survived beyond this timescale ($\approx 35$ Myr) varied by at most about 60%. These results help affirm that collisions occur too quickly to explain WD pollution in the planar CRTBP $2$:$1$ regime, and highlight the need for further periodic orbit studies with the eccentric and inclined TBP architectures and other significant orbital period commensurabilities.