Abstract
Accurate flood simulations are essential for effective prevention but they can be computationally slow and expensive, especially in large-scale scenarios. This can limit their use in time-critical situations. To solve this problem, this study aims to enhance the speed of numerical methods while maintaining accuracy in flood simulation results. A parallel algorithm is developed by applying the automatic domain updating method to solve the shallow water equations using a well-balanced, positivity-preserving first-order finite volume scheme. The parallel algorithm was designed with an index array to store the coordinates of cells in the computational domain that exclude unnecessary cells. The index array is divided and assigned to different cores, enabling parallel processing of each sub-domain. The developed parallel program was tested by simulating the water flow, compared with the results obtained in the literature, and applied to the Xe-Pian Xe-Namnoy dam break simulation in Laos. The computational times obtained by the proposed parallel program were compared with those of the serial program, which used only the automatic domain updating method without the parallel technique. The results show that the parallel program outperforms the serial program by reducing the computational time.