Abstract. This paper analyzes a number of high-order and optimized finite-difference methods for numerically simulating the propagation and scattering of linear waves, such as electromagnetic, acoustic, and elastic waves. The spatial operators analyzed include compact schemes, noncompact schemes, schemes on staggered grids, and schemes which are optimized to produce specific characteristics. The time-marching methods include Runge-Kutta methods, Adams-Bashforth methods, and the leapfrog method. In addition, the following fully-discrete finite-difference methods are studied: a one-step implicit scheme with a three-point spatial stencil, a one-step explicit scheme with a five-point spatial stencil, and a two-step explicit scheme with a five-point spatial stencil. For each method, the number of grid points per wavelength required for accurate simulation of wave propagation over large distances is presented. The results provide a clear understanding of the relative merits of the methods compared, especially the trade-offs associated with the use of optimized methods. A numerical example is given which shows that the benefits of an optimized scheme can be small if the waveform has broad spectral content.Key words. finite-difference methods, wave propagation, electromagnetics, acoustics, Maxwell equations AMS subject classifications. 65M05, 78A40, 78M20, 76Q05PII. S1064827599350320Introduction. Numerical simulation can play an important role in the context of engineering design and in improving our understanding of complex systems. The simulation of wave phenomena, including electromagnetic, elastic, and acoustic waves, is an area of active research. Lighthill [24] and Taflove [40] discuss the prospects for computational aeroacoustics and electromagnetics, respectively. The computational requirements for accurate simulations of the propagation and scattering of waves can be high, particularly if the size of the geometry under study is much larger than the wavelength. Consequently, there has been considerable recent effort directed towards improving the efficiency of numerical methods for simulating wave phenomena.In electromagnetics, the most popular approach to the numerical solution of the time-domain Maxwell equations for numerous applications has been the algorithm of Yee [50], which was named the finite-difference time-domain (FDTD) method by Taflove [41]. This algorithm combines second-order centered differences on a staggered grid in space with second-order staggered leapfrog time marching. Its main attributes are its very low cost per grid node and lack of dissipative error. Yee's method is often applied using Cartesian grids, with a special treatment of curved boundaries [20]. Extension to curvilinear grids was carried out by Fusco [11]. Madsen and Ziolowski [28] put the method into a finite-volume framework applicable to unstructured grids. Vinokur and Yarrow [47,48] developed a related finite-surface method with advantages at boundaries and grid singularities.Other methods which have been successfully app...