Estimate dispersion from Ambient Noise Cross-Correlations using Aki's spectral formulation

This software estimates interstation dispersion for Love and Rayleigh surface waves using Aki's spectral formulation. The method is detailed in the publication:

Hawkins R. and Sambridge M., An adjoint technique for estimation of interstation phase and group dispersion from ambient noise cross-correlations, BSSA, 2019

This software is scientific software and is intended to demonstrate the functionality outlined in the manuscript reference above.

Rhys Hawkins ([email protected])

This code is released under a GPL v3 License.

A portion of the public domain LAPACK codes obtained from netlib.org are included in this repository as a convenience under the forwardmodel/dggev subdirectory as is licensed by the authors under a modified BSD license (see http://www.netlib.org/lapack/LICENSE.txt).

• g++/gfortran or equivalent (versions 8.2.x were used during development)
• FFTW http://www.fftw.org/
• GSL https://www.gnu.org/software/gsl/

• Python 2.7/3
• Matplotlib
• Numpy

You may need to edit the Makefiles in Reference, InitialPhase/optimizer and Phase/optimizer to adjust compiler used.

The top level Makefile compiles all the necessary components so assuming the prerequisites are available as listed above, compilation should be a matter of issuing a `make' command from the top level directory.

Individual components can be built in-order using the following commands if changes are required by your build environment

make -C forwardmodel/dggev
make -C forwardmodel/spec1d
make -C Reference
make -C InitialPhase/optimizer
make -C Phase/optimizer

In the tutorial directory, there are a series of numbered bash scripts for running an example processing of an ambient noise cross-correlation, namely

• 00_create_reference.sh
• 01_create_initial_target_phase.sh
• 02_fit_initial_target_phase.sh
• 03_fit_bessel.sh

There are corresponding scripts for running using only Rayleigh wave observations.

• 01_create_initial_target_phase_rayleigh.sh
• 02_fit_initial_target_phase_rayleigh.sh
• 03_fit_bessel_rayleigh.sh

Within these scripts, you may need to modify the python interpreter used in the 01_create_*.sh scripts to suit.

Additionally, there is a latex document, tutorial.tex which describes in detail the inputs and parameters that are used to control the processing of ambient noise cross correlations with this method.