Models and numerical methods of the impact of tsunamis on coastal forests are of vital im- portance for exploring the potential of coastal vegetation as a means of mitigation. Such a model is formulated as a multilayer shallow water system based on a free-surface formulation of the Euler equations for an ideal fluid. Specifically, the Euler equations are approximated by a layer averaged non-hydrostatic (LDNH) approach involving linear pressures and piece- wise constant velocities. Furthermore, drag forces, inertia forces, and porosity are added to model the interaction with the forest. These ingredients are specified in a layer-wise man- ner. Thus, the vertical features of the forest are described with higher accuracy than within a single-layer approach. Projection methods for the non-hydrostatic pressure in conjunction with polynomial viscosity matrix finite volume methods are employed for the numerical solution of the multilayer model, that is for the propagation of tsunamis and coastal flooding. Experi- mental observations and field data are used to validate the model. In general good agreement is obtained. Results indicate, moreover, that coastal vegetation can operate as an efficient natural barrier against coastal hazards and can significantly reduce the effects of tsunamis.