A recent paper outlining our Galerkin BEM approach.

My tectonic and earthquake cycle research is supported by development of elastic boundary element software. This has required investigations into the best approaches for a wide range of computational and mathematical tasks: singular integration of Green’s functions, approximation of farfield Green’s function interactions with the fast multipole method, flexible treatment of equality and inequality constraints, and a big dose of computational geometry. These are a just a few of the components of this effort. This new software will be able to model arbitrary geometries including surface topography, complex data-derived fault geometries, and material boundaries. Models with millions of elements will be possible in two or three dimensions. Exciting!

A visualization of the structure in a boundary element matrix with rotational symmetry.