Problem Statement |
For synthetic study of FWTfull-scale field data inversion of the crustal velocity model for Canterbury region, we use a 1-D the velocity model of 60km (140km X 40km X 20km as the true model. We use a homogenous model 120km X 46km) developed by Lee. et al (2017) as the initial model and a test configuration including 12 receivers (3 X 4, 43 seismic stations (not equally spacing at every 10km) and 6 sources (2 X 3, equally spacing at every 10km) to generate the synthetic observed data for inversion. The wavefields are generated by numerical solutions of the 3-D elastodynamic/ visco-elastodynamic equations according to a specific velocity model and then compared with the synthetic observed data to extract the misfit between the current model with the true model146 earthquake to start the inversion. The observed data are processed including filtering from 0.02Hz to .5Hz, windowing to use only significant part of the signals. The source locations, magnitudes and shapes are determined prior to the inversion. A visco-elastic forward modelling is applied with consideration of the spatial attenuation from the observed data. After a number of inversion iterations, the final inverted model using FWT has matched well with the given 1-D model and the misfit error has significantly decreased. |
Tasks |
Forward modeling of the elastic wave propagation for the 1-D model to create the synthetic observed data.
Inversion: specify the number of iterations, optimization method or optimal step length (if applicable) and follow the steps for doing inversion at one iteration:
|
Schedule |
|
Verification |
Verifying the inverted result using FWT - Adjoint Wavefield method with FWT-Scattering Integral or SASW methods. |
...