LLNL_ToFi

LLNL_ToFi is a small Python program for tomographic filtering of hypothetical seismic mantle structure $v_S$ or $v_P$ using the resolution matrix $R$ of the LLNL-G3D-JPS model by Simmons et al. (2015). The routine LLNL_ToFi.py performs the matrix-vector multiplication $Rm=m'$ to obtain the filtered version $m'$ of the given seismic model $m.$ To be able to perform this operation, $m$ needs to be given in the parametrization of the LLNL-G3D-JPS model.

Original author: Bernhard Schuberth (Geophysics, LMU Munich, Germany, [email protected])
Contributing author: Tom New (EarthByte, School of Geosciences, The University of Sydney, Australia, [email protected])

License

Development

Development of LLNL_ToFi_3 is hosted on GitHub.
Development of the oringal LLNL_ToFi is hosted on the GitLab server of the Leibniz Supercomputing Centre (LRZ) in the bschuberth/LLNL_ToFi repository.

Data

Input data required by LLNL_ToFi.py are located on the LLNL server or by email request to [email protected].

Documentation

• Information on the resolution matrix and the parametrization of the LLNL-G3D-JPS tomographic model can be found in LLNL-G3D-JPS_R_Matrix_TomoFilter_README.pdf

• The LLNL-G3D-JPS model is described in
  Simmons, N. A., Myers, S. C., Johannesson, G., Matzel, E., & Grand, S. P. (2015). Evidence for long-lived subduction of an ancient tectonic plate beneath the southern Indian Ocean. Geophysical Research Letters, 42(21), 9270–9278.
  https://doi.org/10.1002/2015GL066237

• An example of applying the resolution matrix $R$ to a geodynamic model is described in
  Simmons, N. A., Schuberth, B. S. A., Myers, S. C., & Knapp, D. R. (2019). Resolution and covariance of the LLNL-G3D-JPS global seismic tomography model: applications to travel time uncertainty and tomographic filtering of geodynamic models. Geophysical Journal International, 217(3), 1543–1557.
  https://doi.org/10.1093/gji/ggz102

• Running the code:

  1. To run the code, please first get the necessary input data (i.e., the resolution matrix files) from the source given above and put them into the directory ./DATA.

  2. Edit the file model.py such that the dummy function project_model_3D returns the values of your specific seismic velocity model at the given coordinates (radius, lat, lon).

  3. Run the code in a Python 3.x environment use python3 LLNL_ToFi.py for serial processing, or mpirun -n {number of processes} python3 LLNL_Tofi.py for parallel processing.

     Output files will be stored in the directory ./OUTPUT_FILES. Files containing the reparametrized model will be named according to the variable OUTFILE_PARM_PREFIX [default: LLNL_G3D_JPS_Parm_layer] and the tomographically filtered model will be stored according to the variable OUTFILE_FILT_PREFIX [default: LLNL_G3D_JPS_ToFi_layer].

     Specify the option -n|--no-reparam if you run the code the several times and you do not want to perform the reparametrization again. This assumes that the reparametrized version of your seismic model (i.e., on the parametrization of the LLNL-G3D-JPS tomographic model) is already stored in the directory ./OUTPUT_FILES.