Bachelorarbeiten bei Prof. Mertsch in 2022

 

Cosmic rays on GPUs

Context:
The universe is pervaded by a background of high-energy particles. These cosmic rays scatter with turbulent magnetic fields, both inside their sources and on their way from the sources to the observer. Solving the long-standing problem of cosmic ray origin requires a reliable understanding of these interactions. Yet, only in simplified setups can the distribution of cosmic rays in position and momentum be computed analytically. Specifically, the anisotropic nature of turbulence renders previous numerical simulations questionable.

Goals:
We will attempt to numerically compute the transport of cosmic rays in anisotropic turbulence. To this end, we will set up a turbulent magnetic field on a computer, making extensive use of GPUs. We can then integrate the equations of motion. From the simulation data, we will compute the diffusion coefficient and compare with the standard prediction. We will derive some generic predictions for observations.

Requirements:
Some basic familiarity with C++ will be required, but no experience with CUDA or GPUs is necessary.

A galactic flux of high-energy neutrinos

Context:
The IceCube experiment at the South Pole has been detecting a flux of cosmic high-energy neutrinos. While the sources have not been unambiguously identified, a fraction must be of galactic origin. A detection of this galactic flux would be a major breakthrough in astrophysics. Model predictions are urgently required for searching for this galactic flux, but are notoriously unreliable since a number of unknowns enter the computation.

Goals:
Adopting a multi-messenger approach will result in tighter predictions for galactic neutrinos. We will be using and extending existing codes for computing the fluxes of high energy neutrinos, gamma-rays and cosmic rays from the Galaxy. Comparing to existing data, we will constrain the most important model parameters. In the end we will make a prediction for what IceCube will observe and quantify the uncertainty of this prediction.

Requirements:
Familiarity with python and an interest to learn about high-energy astrophysics.

The ionisation puzzle and a statistical model of molecular clouds

Context:
Cosmic rays are the only particles that can ionise the dense cores of molecular clouds. Yet, the predictions from the models are significantly smaller than the observations. In addition, there is a large scatter in the rate of ionisation observed from molecular clouds that are otherwise comparable in properties. What such models ignore, however, are spatial correlations between clouds and the sources of cosmic rays.

Goals:
We will build a statistical model of the line fluxes from a population of molecular clouds. This will build on a probabilistic model of the cosmic ray flux in the Galaxy and our understanding of the transport of cosmic rays into molecular clouds. In the end, we will either be able to solve the ionisation puzzle or conclude that more substantial modifications of the ecology of the Galaxy are required.

Requirements:
This model has the largest analytical part, so only limited coding experience is required.