Celestial Mechanics and Dynamical Astronomy

, Volume 107, Issue 4, pp 487–500

Orbital stability of systems of closely-spaced planets, II: configurations with coorbital planets


    • Deptartment of Mechanical EngineeringStanford University
  • Jack J. Lissauer
    • Space Science and Astrobiology Division, MS 245-3National Aeronautics and Space Administration, Ames Research Center
Original Article

DOI: 10.1007/s10569-010-9288-0

Cite this article as:
Smith, A.W. & Lissauer, J.J. Celest Mech Dyn Astr (2010) 107: 487. doi:10.1007/s10569-010-9288-0


We numerically investigate the stability of systems of 1 \({{\rm M}_{\oplus}}\) planets orbiting a solar-mass star. The systems studied have either 2 or 42 planets per occupied semimajor axis, for a total of 6, 10, 126, or 210 planets, and the planets were started on coplanar, circular orbits with the semimajor axes of the innermost planets at 1 AU. For systems with two planets per occupied orbit, the longitudinal initial locations of planets on a given orbit were separated by either 60° (Trojan planets) or 180°. With 42 planets per semimajor axis, initial longitudes were uniformly spaced. The ratio of the semimajor axes of consecutive coorbital groups in each system was approximately uniform. The instability time for a system was taken to be the first time at which the orbits of two planets with different initial orbital distances crossed. Simulations spanned virtual times of up to 1 × 108, 5 × 105, and 2 × 105 years for the 6- and 10-planet, 126-planet, and 210-planet systems, respectively. Our results show that, for a given class of system (e.g., five pairs of Trojan planets orbiting in the same direction), the relationship between orbit crossing times and planetary spacing is well fit by the functional form log(tc/t0) = bβ + c, where tc is the crossing time, t0 = 1 year, β is the separation in initial orbital semimajor axis (in terms of the mutual Hill radii of the planets), and b and c are fitting constants. The same functional form was observed in the previous studies of single planets on nested orbits (Smith and Lissauer 2009). Pairs of Trojan planets are more stable than pairs initially separated by 180°. Systems with retrograde planets (i.e., some planets orbiting in the opposite sense from others) can be packed substantially more closely than can systems with all planets orbiting in the same sense. To have the same characteristic lifetime, systems with 2 or 42 planets per orbit typically need to have about 1.5 or 2 times the orbital separation as orbits occupied by single planets, respectively.


StabilityPlanetary systemsCoorbital planets

Copyright information

© Springer Science+Business Media B.V. 2010