Abstract
This paper presents a 2D model for predicting tsunami propagation on dynamically adaptive grids. In this model, a finite volume Godunov-type scheme is implemented to solve the 2D nonlinear shallow water equations on adaptive grids. The simplified adaptive grid achieves automatic adaptation through increasing or reducing the subdivision level of a background cell according to certain criteria defined by tsunami wave features. The grid system is straightforward to implement and no data structure is needed to store grid information. The present model is validated by applying it to simulate three laboratory-scale test cases of tsunami propagation over uneven beds and finally used to reproduce the 2011 Tohoku tsunami in Japan. The model results confirm the model's capability in predicting tsunami wave propagation in a reliable and efficient way.