A weighted cellular automata 2D inundation model for rapid flood analysis

Abstract

To achieve fast flood modelling for large scale problems, a two-dimensional cellular automata based model has been developed in this study. This model uses simple transition rules and a weight-based system instead of solving complex shallow water equations. The cellular automata feature allows the proposed model to be implemented in a parallel computing environment such that the modelling efficiency is improved significantly due to the combination of simplification and parallelisation. The proposed model has been tested on four hypothetical case studies and one real world example, and the outputs compared to those from traditional physically based hydraulic models. Results show that the proposed model is capable of producing good agreement with other hydraulic models, using a fraction of computational time. In the case of the real world example, the model run times are up to 8 times faster than the a widely use commercial hydraulic model. This rapid and accurate attributes of the proposed model have demonstrated its applicability for quick flood analysis in large modelling systems