Computational Astrophysics (PHYS11037)
Physics and Astronomy
Normal Year Taken
Delivery Session Year
Practical programming experience.
This course provides an introduction to advanced computational techniques used for numerical simulations in astrophysics involving gravity and/or fluids. The topics include N-body methods for solving gravity problems and numerical hydrodynamics techniques for fluids.Astrophysical topics for which the methods are used include cosmological simulations of structure formation in the Universe, the formation and evolution of galaxies, the formation and evolution of stars and planetary systems, and the collisions of neutron stars and black holes as a model for gamma-ray bursters. For more information on these and other topics to which the methods are applied, please see: http://www.roe.ac.uk/~aam/ecca.Although the examples are drawn from astrophysics, the methods taught are applicable to a wide range of problems in computational physics. The course is continuously assessed on the basis of workshop exercises: there is no Degree Examination.N.B. This is a course on numerical algorithms and their practical use, not a course that teaches programming techniques or languages. Students must already be competent programmers.
The course teaches students:- how to formulate the equations relevant for hydrodynamics in conservative and non-conservative form.- how to discretise the equations relevant for hydrodynamics in conservative form.- how to numerically implement as computer code a subset of the equations relevant for hydrodynamics.- knowledge of concepts of source terms, Eulerian and Lagrangian formulations, implicit and explicit formulations, finite difference approximations, finite difference/volume/element methods.- the Smoothed Particle Hydrodynamics (SPH) implementation of the hydrodynamics equations.- an understanding of the situations in which a Lagrangian treatment (as used by SPH) may be more appropriate than a Eulerian treatment.- the Particle-Mesh method of solving the Poisson equation. This includes the ability to express the equations for gravitational dynamics in Fourier space, a sufficient awareness of a representative code to alter the initial conditions appropriately, understand the output, check the accuracy of the results, and manipulate and display them using standard tools.- how to numerically implement as computer code a subset of the equations relevant for Particle-Mesh simulations.- properties of the Discrete Fourier Transform.
Written Exam 0%, Coursework 0%, Practical Exam 100%
All course information obtained from this visiting student course finder should be regarded as provisional. We cannot guarantee that places will be available for any particular course. For more information, please see the visiting student disclaimer: