Nemesis: An Optimised Algorithm for Simulating Planetary Systems in Clusters

Abstract

We introduce a revamped Nemesis, a numerical tool designed to circumvent the multi-scale, multi-physics problem inherent to numerical simulations of planetary systems in star clusters. We validate the algorithm by comparing it against both a 4th-order Hermite and a symplectic N-body code in simulations of a star cluster hosting five planetary systems, including ~1000 asteroids. Nemesis finds indistinguishable orbital parameters of evolved planetary systems when compared to its direct N-body code counterparts. Moreover, energy errors are comparable to symplectic codes and the algorithm reduces computation time by up to several orders of magnitude. Due to its parallelisation, t, the performance gain increases significantly with the number of planetary systems. As a result, preliminary results of an ongoing simulation consisting of ~4600 star cluster (~4500 planetary systems) is shown, focusing on captured planets, rogue planets and collisional events.

Authors 2

1. Erwan Hochart Aachen Sterrewacht Leiden

   Leiden University · Leiden Observatory

   Affiliation as printed

   Sterrewacht Leiden, Leiden University

2. Simon Portegies Zwart corresponding Aachen Sterrewacht Leiden

   Leiden University · Leiden Observatory

   Affiliation as printed

   Sterrewacht Leiden, Leiden University