Functionally graded materials (FGMs) have spatial variations in composition and structure resulting in corresponding changes in material properties. In this work, a model for the laser generation of ultrasound in FGMs is used to explore the dependence of the generated acoustic waves on the compositional profile, and to evaluate the feasibility of using laser generated surface acoustic waves (SAWs) to determine the depth dependent properties of graded coating materials. The materials studied are coatings which exhibit elastic property gradations from the coating surface to the interface with the substrate. The graded coatings are modeled as multilayer systems and the number of layers required to achieve a desired level of convergence of the model over a given frequency range is determined. For an assumed compositional profile, the elastic properties of the respective layers are calculated using a modified rule of mixtures formulation. Numerical results showing the sensitivity of SAW dispersion behavior to various compositional profiles are presented. In addition, an inversion algorithm is developed to extract position dependent elastic properties from measured SAW dispersion behavior and preliminary results are obtained using simulated experimental data. © 2004 American Institute of Physics