A new numerical and analytical scheme to solve the full wave equation for seismic modeling based on REM and Leapfrog methods

Document Type : Research Article

Authors

1 Faculty of Mining, Petroluem and Geophysivs, Shahrood University of Technology, Shahrood, Iran.

2 Currently: Department of Mining Engineering, College of Engineering, University of Tehran, Tehran, Iran; Attendant Professor, Faculty of Mining, Petroleum and Geophysics, Shahrood University of Technology, Shahrood, Iran.

3 Center for Research in Geophysics and Geology (CPGG), Federal University of Bahia (UFBA), Salvador, Brazil

4 Faculty of Mining, Petroleum and Geophysics, Shahrood University of Technology, Shahrood, Iran

Abstract

Summary
The aim of this paper is to present a new numerical method to solve the wave equation with a good accuracy and high stability using the Leapfrog symplectic integrator and rapid expansion method (REM). It can be used for seismic modeling and reverse time migration (RTM). Using the REM with Fourier transform method for spatial derivative, the Leapfrog-rapid expansion method (L-REM) can be even used for larger time steps. The L-REM provides the solution of the wave equation and its first time derivative at the current time step. In addition to the very low error for the small time steps, increasing the time step also lead to  more accurate results and high stability in comparison with the similar methods such as Störmer-Verlet, Leapfrog and Störmer-Verlet-rapid expansion method which will also be discussed in this paper.
 
Introduction
Wave-field extrapolation is implemented by solving the wave equation through various mathematical methods. The finite difference method is a well-known and popular numerical tool to discretize the wave equation, and its use has been common in the approximation of the spatial and time derivatives for a wave-field. Originally, the time operator was approximated by a second-order scheme, whereas the spatial derivatives were approximated by a fourth-order scheme. Approximating the time derivative in this way may introduce numerical error, leading to distortion of the pulse and numerical dispersion, which can be avoided with small time steps at the expense of increasing the computational time. Also some other methods such as Störmer-Verlet (SV), Leapfrog (L) and Störmer-Verlet-rapid expansion method (SV-REM) have been presented to improve the solution of wave equation. In the current study, a symplectic scheme based on Leapfrog integrator and rapid expansion method (L-REM) is proposed to extrapolate the wave-field and its first derivatives in time for the same time step which can be used to calculate Poynting vectors for wave-field separation and to calculate the reflection angles.
 
Methodology and Approaches
In order to verify the numerical accuracy and behaviour of the error associated with Leapfrog-REM scheme a numerical example has been presented to be solved using different time sampling values. For implementation, an explosive source used in the centre of the computational domain having a Ricker wavelet with a maximum frequency of 25 Hz.
 
Results and Conclusions
The presented L-REM scheme provides the solution of the wave equation and its first time derivative for different time steps. In addition to the very low error for the small time steps, increasing the time step also lead in the more accurate results and high stability in comparison with the similar methods.

Keywords

Main Subjects

References

[1] Etgen, J. (1986). High-order finite-difference reverse time migration with the 2-way non-reflecting wave equation. Stanford Exploration Project, report SEP-48, 133–146.
[2] Kosloff, D., Filho, A., Tessmer, E., & Behle A. (1989). Numerical solution of the acoustic and elastic wave equation by new rapid expansion method. Geophysical Prospecting, (37), 383–394.
[3] Pestana, R. C., & Stoffa, P. L. (2010). Time evolution of the wave equation using rapid expansion method. Geophysics, 75(4), T121-T131.
[4] Soubaras, R., & Zhang, Y. (2008). Two-step explicit marching method for reverse time migration. 70th Annual International Conference and Exhibition, EAGE, Extended Abstracts.
[5] Chen, J. M. (2007). High-order time discretizations in seismic modelling. Geophysics, 72(5), SM115–SM122.
[6] Dablain, M. A. (1986). The application of high-order differencing to the scalar wave equation. Geophysics, 51, 54–66.
[7] Zhang, Y., & and Zhang, G. (2009). One-step extrapolation method for reverse time migration. Geophysics, 74( 4), A29–A33.
[8] Schmidta, K., Diazb, J. & Heiera, C. (2015). Non-conforming Galerkin finite element methods for local absorbing boundary conditions of higher order.  Computers & Mathematics with Applications, 70(9), 2252-2269.
[9] Stanglmeiera, M., Nguyena, N.C., Perairea J. & Cockburnb, B. (2016). An explicit hybridisable discontinuous Galerkin method for the acoustic wave equation. Computer Methods in Applied Mechanics and Engineering, 300,748–769.
[10] Baldassari, C., Barucq, H., Calandra, H., Denel, B., Diaz, J. (2009). Ultrasonic Wave Propagation in Non Homogeneous Media: The reverse time migration technique coupled with finite element methods. Springer Berlin Heidelberg, pp 207-216.
[11] Tal-Ezer, H., Kosloff, D., & Koren Z. (1987). An accurate scheme for forward seismic modelling. Geophysical Prospecting, (35), 479–490.
[12] Tessmer, E. (2011). Using the rapid expansion method for accurate time-stepping in modeling and reverse-time migration. Geophysics, 76(4), S177–S185.
[13] Skell, R. H., Zhang G., & Schlick, T. (1997). A family of symplectic integrators: Stability, accuracy, and molecular dynamics applications. SIAM Journal on Numerical Analysis, (18), 203–222.
[14] Bonomi, E., Brieger L., Nardone, C., & Pieroni, E. (1998). 3D spectral reverse time migration with no-wraparound absorbing conditions. 78th Annual International Meeting, SEG, Expanded Abstracts, 1925–1928.
[15] Arnold, V. I. (1989). Mathematical methods of classical mechanics (2nd ed.). 60, Springer.
[16] Yoshida, H. (1990). Construction of higher order symplectic integrators. Physics Letters A, 150, 262–268.
[17] Sexton, J. C., &  Weingarten D. H. 1992. Hamiltonian evolution for the hybrid Monte Carlo algorithm. Nuclear Physics B, 380(3):665.
[18]Araujo, E. S., Pestana, C. R., & dos Santos, A. W. G. (2013). Symplectic schmes and poyting vector in the reverse time migration. 83rd Annual International Meeting, SEG, Expanded Abstracts.

Files

History

  • Receive Date: 18 May 2015
  • Revise Date: 02 February 2016
  • Accept Date: 28 February 2016

Share

How to cite

Statistics