SMIwiz: An integrated toolbox for multidimensional seismic modelling and imaging

This paper contributes an open source software SMIwiz, which integrates seismic modelling, reverse time migration (RTM), and full waveform inversion (FWI) into a unified computer implementation. SMIwiz has the machinery to do both 2D and 3D simulation in a consistent manner. The package features a number of computational recipes for efficient calculation of imaging condition and inversion gradient: a dynamic evolving computing box to limit the simulation cube and a well-designed wavefield reconstruction strategy to reduce the memory consumption when dealing with 3D problems. The modelling in SMIwiz runs independently: each shot corresponds to one processor in a bijective manner to maximize the scalability. A batchwise job scheduling strategy is designed to handle large 3D imaging tasks on computer with limited number of cores. The viability of SMIwiz is demonstrated by a number of applications on benchmark models. Program summary Program Title: SMIwiz CPC Library link to program files: https://doi .org /10 .17632 /tygszns27k .1 Developer's repository link: https://github .com /yangpl /SMIwiz Licensing provisions: GNU General Public License v3.0 Programming language: C, Shell, Fortran External dependencies: MPI [1], FFTW [2] Nature of problem: Seismic modelling and imaging (FWI and RTM) Solution method: High-order finite-difference time-domain (FDTD) for modelling on staggered grid; Quasi Newton LBFGS algorithm for nonlinear optimization; line search to estimate step length based on Wolfe condition References [1] https://www .mpich .org/ [2] http://fftw .org/

Automated summary

This paper presents an open source software called SMIwiz, which combines seismic modelling, reverse time migration, and full waveform inversion into a unified computer implementation. SMIwiz supports both 2D and 3D simulations and provides various computational recipes for efficient calculation. Its independent processing and batchwise job scheduling ensure scalability, and its viability is demonstrated through applications on benchmark models.