Planets form in discs of gas and dust around young stars. Within these discs, micron-sized dust particles need to clump together to grow 14 orders of magnitude to form Earth-like planets as well as the cores of giant planets. It is a major challenge to understand dust growth from start to finish. State of the art observations provide spectacular glimpses of the dust distribution at a limited range of sizes: ALMA produces images of the thermal emission of mm-sized dust, while instruments such as SPHERE probe the distribution of much smaller particles. However, for a comprehensive theory of planet formation, we need to understand the process from start to finish, from micron-sized to planet-sized. This is therefore the story of the dust size distribution: how many dust specks, pebbles and boulders are present? While there are large size ranges that are out of reach observationally, in this project we will exploit the fact that all dust sizes are coupled to the gas via friction to take a panoptic view of the size distribution for the first time. Since the gas feels friction from all dust sizes, the size distribution is encoded in the gas kinematics, and therefore in every single dust size as well. We will perform hydrodynamical simulations including the full dust size distribution to write the polydisperse story of planet formation. We aim to reconstruct the full size distribution from sparse observations, thereby avoiding the need for expensize multi-wavelength observations. We will compare dust and gas distributions with observations of protoplanetary discs as well as the composition of Solar system bodies. We will use a novel numerical method that allows us to perform these computationally expensive simulations, and employ machine learning to speed up the calculations. This way, we will for the first time be able to build up a complete picture of how dust particles grow into planets and construct a comprehensive model of planet formation.
Fields of science
- natural sciencesphysical sciencesastronomyplanetary sciencesplanetsgiant planets
- natural sciencesphysical sciencesastronomyplanetary sciencesplanetsexoplanetology
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
- natural sciencesmathematicsapplied mathematicsnumerical analysis
- HORIZON.1.1 - European Research Council (ERC) Main Programme