Thesis Topic: Exploring the physics of cosmic dust by Monte Carlo radiative transfer simulations of polarised light
Thesis Supervisors: Ralf Siebenmorgen and Mario Van Den Ancker
During their travel through the cosmos from stars to us, photons are scattered, polarised, absorbed, and re-emitted by dust particles. Dust surrounds various astrophyiscal objects such as young stellar objects, proto-planetary or debris disks, supernovae, starburst galaxies, active galactic nuclei, and gamma ray bursts. Dust is an important component in the general field of interstellar and extra-galactic medium, the full characterisation of which is of crucial importance to understand various fundamental astrophysical processes.
Polarized light is a relative unexplored window to the universe and its study will boost our understanding of various types. Comparison of continuum polarization in the optical and submillimeter continuum increased substantially our understanding of the composition and structure of cosmic dust. Polarimetric continuum observations with optical/IR instruments as available by FORS and SPHERE can be combined with those obtained in the submillimeter with Planck and ALMA. A physical interpretation of such modern data sets requires state-of-the- art polarized light models.
Recent advances in computer technology enables the use of efficient Monte Carlo techniques that are parallelized on GPU or CPUs allowing to study the flight path of photons in arbitrary three-dimensional dusty media.Our tMCpol code includes treatment of absorption, scattering, and emission by large grains and quantum heated nanoparticles (PAHs) as well as polarized light with full treatment of the Stokes formalism and mechanisms of dichroic polarisation and birefringence due to absorption and emission of photons by non-spherical and (magnetically) aligned dust particles. Benchmark comparison between different of such codes shows a large scatter with tMCpol as the only one available that shows high precision against analytical solutions. It outputs the radiation from X-rays to mm wavelengths and produces high resolution images.
Only such physical models will enable detailed analysis of existing dust polarization data for which recent ALMA observations of proto-planetary disks or SPHERE imaging of the hyper- luminous supergiant VYCMa are serving as examples for the project. The present step in the continued development of tMCpol is to include time dependency for future treatment of transient astrophysical phenomena such as SN and GRB.
The goal of this PhD project is to study and explore the physics of the dust by means of the Monte Carlo radiative dust polarization code, tMCpol and confront model predictions against available polarimetric observations. This is a timely project to support understanding of new data obtained with our modern instrumentation that are available at the VLT, ALMA, and in space with the GAIA and Planck mission, and/or to simulate updated (non-polarimetric) science cases for METIS at ESO's Extremely Large Telescope (ELT).
Key competences that will be learned during the PhD project include performing advanced numerical simulations, programming techniques for vectorized code on GPU, reducing, and analyzingdatasets from state-of-the-art optical and radio telescopes,scientific analysis, writing scientific papers, and writing observing proposals.