PythonMHD: A New Code For Astrophysical MHD Simulations

Nov 14, 2023 - 10:40 am to 11:30 am
Location

Campus, PAB 102/103

Speaker
Delica Leboe-McGowan (University of Manitoba) In Person and zoom https://stanford.zoom.us/j/98604058568

Zoom info: https://stanford.zoom.us/j/98604058568

PythonMHD is a new software package for astrophysical magnetohydrodynamic (MHD) simulations. It is the first comprehensive, research-oriented MHD simulation code to be written in Python. By using an accessible, widely understood programming language, PythonMHD minimizes its learning curve for new users, allowing more individuals to explore astrophysical MHD phenomena. In order to further improve the user’s experience, PythonMHD provides built-in visualization and analysis tools that the user can access while they are still running their simulation. The inclusion of these visualization and analysis tools is particularly important for preventing users from wasting time and computational resources on unusable simulation data, because they can terminate a simulation as soon as they see unexpected results in the visualization or analysis outputs. Although Python is ideally suited for creating a readable, user-friendly MHD code, its speed compared to low-level languages, such as C, has prevented it from being considered as a serious candidate for computationally intensive MHD algorithms. In order to address this critical problem, PythonMHD gives the option of running its MHD solver in C, enabling users to efficiently run large simulations while benefitting from the visualization and analysis functions that a Python-based tool can more easily provide. Using both standard test problems and original simulations, this talk will explore how PythonMHD can help to accelerate novel research on the behaviour of magnetized gas in the interstellar medium.