Crustal Deformation and Fault Mechanics Paul Segall

Software

hmmvp

Construct an H-matrix and compute matrix-vector products of the form B*x, B(rs,:)*x, and B(rs,cs)*x(cs). An H-matrix is one way to compress the matrix relating slip to stress in a BEM simulation of a fault. The package hmmvpex contains practical usage examples. (Previous major version.)

Download hmmvp

dc3dm

Construct and apply an H-matrix approximation to the displacement-discontinuity method (DDM) linear operator relating quasistatic dislocation and traction components on a nonuniformly discretized rectangular fault in a homogeneous elastic half space. More information can be found on this poster presented by Andrew Bradley during the 2013 AGU Fall Meeting.  

Download dc3dm

kfgs

Use the adjoint method to compute the gradient of the log likelihood function from a Kalman filter. The gradient calculation takes about as long as running the Kalman filter once independent of the number of parameters.

disloc3d

disloc3d is a Matlab wrapper to Y. Okada's DC3.F. It uses geometry conventions specific to our group. There are three versions of the routine: 1. a pure Matlab translation (modified from Coulomb 3.1); 2. a basic Matlab mex wrapper; and 3. a version that (a) parallelizes the calculations using OpenMP and (b) compensates for numerical error that occurs along rays extending from a dislocation element's edges.