How did the Earth and the other terrestrial planets form? Even though there are many models in existence that attempt to solve this question, all of them suffer from specific shortages that can only be overcome with next-generation computation facilities.
At present the rapid formation of Mars – which has been constrained to occur within 5 Myr from hafnium-tungsten and iron-nickel chronology of martian meteorites – can only be reproduced with simulations on GPUs starting with tens of thousands of planetesimals.
These models assume, however, that the gas giants Jupiter and Saturn remained on their current orbits with their current masses. Yet, chronology of meteorites believed to originate in the outer solar system (OSS) has shown that these planets formed in 1-3 Myr, concurrently with Mars.
As the gas giants formed they scattered material from the OSS into the inner solar system (ISS). OSS material has a different isotopic and chemical composition than ISS material, so that the growth of the giant planets is thought to have polluted the ISS. Monte Carlo mixing models using isotopes of various meteorite classes show that Earth and Mars can at most incorporate 10% of OSS material.
Here we want to employ high-resolution GPU simulations to quantify how much OSS material may end up in the ISS and compare our models with cosmochemical data and employ Monte Carlo mixing models for the isotopic composition of the Earth and Mars. The number of free parameters in the GPU simulations is large here we restrict ourselves to only vary the diameter of the planetesimals and the growth rate of the giant planets. We aim to run 64 GPU simulations starting with 32k self-gravitating planetesimals.
In the absence of self-consistent models for the formation of the gas giants, we model the growth of Jupiter and Saturn based on an analytical prescription. We simulate the growth of the terrestrial planets from planetesimal accretion simultaneously with that of the giant planets.
The main scientific and sociological advance is that the planetary science community is one step closer to unravelling the dynamical history of the early solar system and understanding how the planets formed. The planetary science community knows very little about the formation of the giant planets and it is mostly unexplored in the literature. The main technical advance is progress in parallelisation algorithms specific to planet formation and planetary dynamics/celestial mechanics.
University of Bayreuth, Germany