Skip to main content
Log in

Acoustic Wave Propagation in Complicated Geometries and Heterogeneous Media

  • Published:
Journal of Scientific Computing Aims and scope Submit manuscript

Abstract

We construct finite difference discretizations of the acoustic wave equation in complicated geometries and heterogeneous media. Particular emphasis is placed on the accurate treatment of interfaces at which the underlying media parameters have jump discontinuities. Discontinuous media is treated by subdividing the domain into blocks with continuous media. The equation on each block is then discretized with finite difference operators satisfying a summation-by-parts property and patched together via the simultaneous approximation term method. The energy method is used to estimate a semi-norm of the numerical solution in terms of data, showing that the discretization is stable. Numerical experiments in two and three spatial dimensions verifies the accuracy and stability properties of the schemes.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11

Similar content being viewed by others

References

  1. Appelö, D., Petersson, N.A.: A fourth-order accurate embedded boundary method for the wave equation. SIAM J. Sci. Comput. 34(6), 2982–3008 (2012)

    Article  MathSciNet  Google Scholar 

  2. Browning, G., Kreiss, H.-O., Oliger, J.: Mesh refinement. Math. Comput. 27, 29–39 (1973)

    Article  MATH  MathSciNet  Google Scholar 

  3. Carpenter, M., Gottlieb, D., Abarbanel, S.: Time-stable boundary conditions for finite-difference schemes solving hyperbolic systems: methodology and application to high-order compact schemes. J. Comput. Phys. 111, 220–236 (1994)

    Article  MATH  MathSciNet  Google Scholar 

  4. Gustafsson, B., Kreiss, H.-O., Oliger, J.: Time Dependent Problems and Difference Methods. Wiley, London (1995)

    MATH  Google Scholar 

  5. Henshaw, W.D.: A high-order accurate parallel solver for Maxwellś equations on overlapping grids. SIAM J. Sci. Comput. 5(28), 1730–1765 (2006)

    Article  MathSciNet  Google Scholar 

  6. Jensen, F.B., Ferla, C.M.: Numerical solutions of range-dependent benchmark problems in ocean acoustics. J. Acoust. Soc. Am. 87, 1499–1510 (1990)

    Article  Google Scholar 

  7. Jensen, F.B., Kuperman, W.A., Porter, M.B., Schmidt, H.: Computational Ocean Acoustics. Springer, New York (2000)

    Google Scholar 

  8. Knupp, P., Steinberg, S.: Fundamentals of Grid Generation. CRC Press, Boca Raton, FL (1993)

    Google Scholar 

  9. Kreiss, H.-O., Oliger, J.: Comparison of accurate methods for the integration of hyperbolic equations. Tellus 24, 199–215 (1972)

    Article  MathSciNet  Google Scholar 

  10. Kreiss, H.-O., Petersson, N.A.: An embedded boundary method for the wave equation with discontinuous coefficients. SIAM J. Sci. Comput. 28(6), 2054–2074 (2006)

    Article  MATH  MathSciNet  Google Scholar 

  11. Kreiss, H.-O., Petersson, N.A.: A second order accurate embedded boundary method for the wave equation with Dirichlet data. SIAM J. Sci. Comput. 27(4), 1141–1167 (2006)

    Article  MATH  MathSciNet  Google Scholar 

  12. Kreiss, H.-O., Petersson, N.A., Ystrom, J.: Difference approximations of the Neumann problem for the second order wave equation. SIAM J. Numer. Anal. 42(3), 1292–1323 (2004)

    Article  MATH  MathSciNet  Google Scholar 

  13. Kreiss, H.-O., Petersson, N.A.: Boundary estimates for the elastic wave equation in almost incompressible materials. SIAM J. Numer. Anal. 50(3), 1556–1580 (2012)

    Article  MATH  MathSciNet  Google Scholar 

  14. Mattsson, K., Svärd, M., Carpenter, M., Nordström, J.: High order accurate computations for unsteady aeordynamics. Comput. Fluids 36(3), 636–649 (2007)

    Article  MATH  Google Scholar 

  15. Mattsson, K.: Summation by parts operators for finite difference approximations of second-derivatives with variable coefficients. J. Sci. Comput. 51(3), 650–682 (2012)

    Article  MATH  MathSciNet  Google Scholar 

  16. Mattsson, K., Carpenter, M.: Stable and accurate interpolation operators for high-order multiblock finite difference methods. SIAM J. Sci. Comput. 32(4), 2298–2320 (2010)

    Article  MATH  MathSciNet  Google Scholar 

  17. Mattsson, K., Ham, F., Iccarino, G.: Stable boundary treatment for the wave equation on second order form. J. Sci. Comput. 41(3), 366–383 (2009)

    Article  MATH  MathSciNet  Google Scholar 

  18. Mattsson, K., Ham, F., Iaccarino, G.: Stable and accurate wave propagation in discontinuous media. J. Comput. Phys. 227(19), 8753–8767 (2008)

    Google Scholar 

  19. Mattsson, K., Nordström, J.: Summation by parts operators for finite difference approximations of second derivatives. J. Comput. Phys. 199(2), 503–540 (2004)

    Article  MATH  MathSciNet  Google Scholar 

  20. Nissen, A., Kreiss, G., Gerritsen, M.: High order stable finite difference methods for the Schrödinger equation. J. Sci. Comput. 55(1), 173–199 (2013)

    Article  MATH  MathSciNet  Google Scholar 

  21. Nordström, J., Carpenter, M.H.: High-order finite difference methods, multidimensional linear problems and curvilinear coordinates. J. Comput. Phys. 173(26), 149–174 (2001)

    Article  MATH  MathSciNet  Google Scholar 

  22. Nordström, J., Gong, J., Wiede, E., Svärd, M.: A stable and high order multi-block method for the compressible Navier–Stokes equations. J. Comput. Phys. 228, 9020–9035 (2009)

    Article  MATH  MathSciNet  Google Scholar 

  23. Petersson, N.A., Sjögreen, N.: Stable grid refinement and singular source discretization for seismic wave simulations. Commun. Comput. Phys. 8(5), 1074–1110 (2010)

    Google Scholar 

  24. Strand, B.: Summation by parts for finite difference approximations for d/dx. J. Comput. Phys. 110(1), 47–67 (1994)

    Article  MATH  MathSciNet  Google Scholar 

  25. Sturm, F.: Investigation of 3-D benchmark problems in underwater acoustics: a uniform approach. In: Proceedings of the 9th European Conference on Underwater Acoustics vol. 2, pp. 759–764 (2008)

  26. Sturm, F., Ivansson, S., Jiang, Y., Chapman, N.R.: Numerical investigation of out-of-plane sound propagation in a shallow water experiment. J. Acoust. Soc. Am. 124(6), 341–346 (2008)

    Article  Google Scholar 

  27. Virta, K., Duru, K.: High Order Finite Difference Schemes for the Elastic Wave Equation in Discontinuous Media. Available as preprint arXiv:1309.5768 (2013)

  28. Waldén, J.: On the approximation of singular source terms in differential equations. Numer. Methods Partial Differ. Equ. 15(4), 503–520 (1999)

    Article  MATH  Google Scholar 

  29. Whitham, G.B.: Linear and Nonlinear Waves. Wiley, London (1974)

    MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Kristoffer Virta.

Appendices

Appendix 1: Proof of Lemma 2

The proof of Lemma 2 uses the structure of the \(2p\)-th order accurate narrow diagonal variable coefficient second derivative SBP operator constructed in [15]. For reference the operators \(D_1\) and \(D_2^{(b)}\) for the case \(p = 2\) are formulated explicitly. The norm \(H\) is given by

$$\begin{aligned} H = \varDelta x~\mathrm{diag}\left( \frac{17}{48},\frac{59}{48},\frac{43}{48},\frac{49}{48},1,\dots ,1,\frac{49}{48},\frac{43}{48},\frac{59}{48},\frac{17}{48}\right) . \end{aligned}$$

The 4-th order accurate narrow diagonal first derivative SBP operator \(D_1\) is:

$$\begin{aligned} D_1 = \frac{1}{\varDelta x}\begin{bmatrix} \frac{-24}{17}&\quad \frac{59}{34}&\quad \frac{-4}{17}&\quad \frac{-3}{34}&\quad 0&\quad 0\\ \frac{-1}{2}&\quad 0&\quad \frac{1}{2}&\quad 0&\quad 0&\quad 0\\ \frac{4}{43}&\quad \frac{-59}{86}&\quad 0&\quad \frac{59}{86}&\quad \frac{-4}{43}&\quad 0\\ \frac{3}{98}&\quad 0&\quad \frac{-59}{98}&\quad 0&\quad \frac{32}{49}&\quad \frac{-4}{49}\\&\quad&\quad&\quad \frac{1}{12}&\quad \frac{-8}{12}&\quad 0&\quad \frac{8}{12}&\quad \frac{-1}{12}\\&\quad&\quad&\quad&\quad \ddots&\quad \ddots&\quad \ddots&\quad \ddots&\quad \ddots \end{bmatrix} \end{aligned}$$

With the lower right \(4 \times 6\) block obtained by rotating the upper left \(4 \times 6\) block \(180^{\circ }\) and changing the signs of the elements. The part \(M^{(b)}\) of the 4-th order accurate narrow diagonal variable coefficient second derivative SBP operator compatible with \(D_1\) is

$$\begin{aligned} M^{(b)} = D_1^TH B^{(b)} D_1 + \frac{\varDelta x^5}{18} D_3^TC_3 B_3^{(b)} D_3 + \frac{\varDelta x^7}{144} D_4^TC_4 B_4^{(b)}D_4. \end{aligned}$$

Where

$$\begin{aligned} B^{(b)}&= \mathrm{diag}(b_1,\dots ,b_N) = B^{(b)}_4,\\ B^{(b)}_{3}&= \mathrm{diag}\left( \frac{b_1 + b_{2}}{2},\frac{b_i + b_{i+1}}{2},\dots ,\frac{b_{N-1} + b_{N}}{2}\right) ,\\ C_3&= \mathrm{diag}\left( 0,0,\frac{163928591571}{53268010936},\frac{189284}{185893},1,\dots ,1,\frac{189284}{185893},\frac{163928591571}{53268010936},0,0\right) \\ C_4&= \mathrm{diag}\left( 0,0,\frac{1644330}{301051},\frac{156114}{181507},1,\dots ,1,\frac{156114}{181507},\frac{1644330}{301051},0,0\right) , \\ D_3&= \begin{bmatrix} -1&\quad 3&\quad -3&\quad 1&\quad 0&\quad 0 \\ -1&\quad 3&\quad -3&\quad 1&\quad 0&\quad 0\\ d_{1}&\quad d_{2}&\quad d_{3}&\quad d_{4}&\quad d_{5}&\quad d_{6}\\&\quad&\quad -1&\quad 3&\quad -3&\quad 1\\&\quad&\quad&\quad \ddots&\quad \ddots&\quad \ddots&\quad \ddots \end{bmatrix} \end{aligned}$$

with

$$\begin{aligned} d_{1}&= \frac{-185893}{301051}, d_{2} = \frac{79000249461}{54642863857}, d_{3} = \frac{-33235054191}{54642863857},\\ d_{4}&= \frac{-36887526683}{54642863857}, d_{5} = \frac{26183621850}{54642863857}, d_{6} = \frac{-4386}{181507}. \end{aligned}$$

the lower right \(3 \times 6\) block of \(D_3\) is obtained by rotating the upper left \(3 \times 6\) block \(180^{\circ }\) and changing the signs of the elements,

$$\begin{aligned} D_4 = \begin{bmatrix} -1&\quad 3&\quad -3&\quad 1&\quad 0&\quad 0 \\ -1&\quad 3&\quad -3&\quad 1&\quad 0&\quad 0\\ d_{1}&\quad d_{2}&\quad d_{3}&\quad d_{4}&\quad d_{5}&\quad d_{6}\\&\quad&\quad -1&\quad 3&\quad -3&\quad 1\\&\quad&\quad&\quad \ddots&\quad \ddots&\quad \ddots&\quad \ddots \end{bmatrix} \end{aligned}$$

the lower right \(3 \times 6\) block of \(D_4\) is obtained by rotating the upper left \(3 \times 6\) block \(180^{\circ }\). The matrix \(S\) is given by

$$\begin{aligned} S = \frac{1}{\varDelta _x}\begin{bmatrix} -\frac{11}{16}&\quad 3&\quad -\frac{3}{2}&\quad \frac{1}{3}&\quad 0&\quad 0 \\ 0&\quad 1&\quad 0&\quad 0&\quad 0&\quad 0\\ 0&\quad 0&\quad \ddots&\quad 0&\quad 0&\quad 0\\ 0&\quad 0&\quad 0&\quad 0&\quad 1&\quad 0\\ 0&\quad 0&\quad -\frac{1}{3}&\quad \frac{3}{2}&\quad -3&\quad \frac{11}{16} \end{bmatrix}. \end{aligned}$$

1.1 Proof of Lemma 2

Let \(l_p\) be an positive integer, \(b_L\) as in the lemma and \(B_L\) the matrix with the only non-zero elements \(B_{L_{i,i}} = b_L,i = 1\dots l_p\). For \(p = 2\) define

$$\begin{aligned} \begin{array}{l} \tilde{B}^{(b)} = B - B_L,\\ \tilde{B}^{(b)}_{3} = B_3 - B_L,\\ \tilde{B}^{(b)}_{4} = B_4 - B_L,\\ \tilde{M}^{(b)} = D_1^T H \tilde{B} D_1 + \frac{\varDelta x^5}{18} D_3^T C_3 \tilde{B}_3 D_3 + \frac{\varDelta x^7}{144} D_4^T C_4 \tilde{B}_4 D_4,\\ M_L = D_1^T H B_L D_1 + \frac{\varDelta x^5}{18} D_3^T C_3 B_L D_3 + \frac{\varDelta x^7}{144} D_4^T C_4 B_L D_4. \end{array} \end{aligned}$$
(32)

By the choice of \(B_L\) these matrices are symmetric positive semi-definite. Now, \(S^{-T} M_L S^{-1} = \varDelta x b_L \tilde{M}_L\). Where \(\tilde{M}_L\) is independent of \(\varDelta x\) and \(b_L\). By the construction of \(B_L\) the only non-zero elements of \(\tilde{M}_L\) are located in an upper left square block \(A\) of size independent of \(\varDelta x\) but dependent on \(l_p\), by construction \(A = A^T \ge 0\). For \(l_p = 1, \dots \) compute \(A\) and construct the matrix \(B\) with \(B_{ij} = A_{ij}, B_{11} = A_{11} - \beta \). \(\beta \) is then chosen by computing the eigenvalues of \(B\) as the largest number such that \(B \ge 0\). Note that also \(\beta \) is independent of \(\varDelta x\). A choice of \(l_p = 4\) gives a value of \(\beta \) as in Table 1. Let \(\bar{M}_L\) be the matrix resulting from subtracting \(\varDelta x b_L \beta \) from the first diagonal element of \(\varDelta x b_L \tilde{M}_L\). We get

$$\begin{aligned} \mathbf {u}^T M^{(b)} \mathbf {u} = \mathbf {u}^T (\tilde{M}^{(b)} + M_L) \mathbf {u} = \mathbf {u}^T (\tilde{M}^{(b)} + S^T \bar{M}_L S) \mathbf {u} + \frac{\beta \varDelta x}{b_L} \left( \bar{B}^{(b_L)}S\mathbf {u}\right) ^2_1. \end{aligned}$$

Similarly we can derive the second term of (17) by constructing the corresponding \(M_R\) e.t.c., Then taking \(\bar{M}^{(b)} = \tilde{M}^{(b)} + M_L + M_R\) proves the lemma for \(p = 2\). The proof for \(p = 1\) and \(p = 3\) follows the same arguments as for \(p = 2\). \(\square \)

Appendix 2: Dirichlet Boundary Conditions

Assume a homogeneous Dirichlet boundary condition at the west boundary. The corresponding SAT term is then

$$\begin{aligned} SAT_{D_{W}} = \varLambda _\rho ^{-1} H^{-1}_x \left( \tau _{D_W} + \sigma _{D} H_y^{-1} \mathbf {F}^{(-)^T}H_y \right) e_{W} \left( \mathbf {u}_{W} - 0 \right) . \end{aligned}$$
(33)

Similar to the discretization of the interface conditions homogeneous Dirichlet boundary conditions adds a term to the total semi-discrete acoustic energy. The following lemma, proved similarly to Lemma 6, determines the parameter \(\tau _D\) for this term do be positive semi-definite.

Lemma 7

With the same notation as in Lemma 6 and \(\mathbb {1} = \mathrm{diag}(1,\dots ,1)\) the matrix

$$\begin{aligned} \mathcal {T}_D = \begin{bmatrix} -\tau&\quad -\mathbb {1}&\quad -\mathbb {1}\\ -\mathbb {1}&\quad c_1&\quad 0 \\ -\mathbb {1}&\quad 0&\quad c_2 \\ \end{bmatrix} \end{aligned}$$
(34)

is positive semi-definite if \(\tau _j \le -\frac{1}{c_{1_j}} - \frac{1}{c_{2_j}}\).

Define

$$\begin{aligned} \mathbf {U}_D = \begin{bmatrix} \mathbf {u}_{W} \\ (B^{(\rho a_{11})} S_x \mathbf {u})_{W} \\ (\varLambda _\rho \varLambda _{\mathbf {a_{12}}} D_{1 y}\mathbf {u})_{W} \end{bmatrix}, H_D = \begin{bmatrix} H&\quad 0&\quad 0\\ 0&\quad H&\quad 0\\ 0&\quad 0&\quad H \end{bmatrix} \end{aligned}$$
(35)

and let

$$\begin{aligned} \tau _{D_j} \le -\frac{1}{b_ {1R_{j,j}}} -\frac{1}{b_ {2R_{j,j}}}, \tau _D = \mathrm{diag}(\tau _{D_1},\dots ,\tau _{D_{N_y}}) \end{aligned}$$
(36)

where \(b_ {1R_{j,j}}\) and \(b_ {2R_{j,j}}\) are defined by (25), then the addition of the term \(\mathbf {U}_D^T H_D \mathcal {T}_D(\tau _D,b_{1 R},b_{2 R}) \mathbf {U}_D\) to the total semi-discrete energy also defines a semi norm of the solution to (21) and (22). Now with a Dirichlet boundary condition. An energy estimate in this semi-norm follows as in Theorem 1.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Virta, K., Mattsson, K. Acoustic Wave Propagation in Complicated Geometries and Heterogeneous Media. J Sci Comput 61, 90–118 (2014). https://doi.org/10.1007/s10915-014-9817-1

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10915-014-9817-1

Keywords

Navigation