Flood Simulation by a Well-Balanced Finite Volume Method in Tapi River Basin, Thailand, 2017

Abstract

Flood simulation of a region in southern Thailand during January 2017 is presented in this work. The study area covers the Tapi river, the longest river in southern Thailand. The simulation is performed by applying the two-dimensional shallow water model in the presence of strong source terms to the local bottom topography. The model is solved numerically by our finite volume method with well-balanced property and linear reconstruction technique. This technique is accurate and efficient at solving for complex flows in the wet/dry interface problem. Measurements of flows are collected from two gauging stations in the area. The initial conditions are prepared to match the simulated flow to the measurements recorded at the gauging stations. The accuracy of the numerical simulations is demonstrated by comparing the simulated flood area to satellite images from the same period. The results are in good agreement, indicating the suitability of the shallow water model and the presented numerical method for simulating floodplain inundation.

Parallel Algorithms of Well-Balanced and Weighted Average Flux for Shallow Water Model Using CUDA

“…The flux gradient at each cell interface can be approximated by the TVD-WAF (total variation diminishing-weighted average flux) or HLL (Harten-Lax-van Leer) methods [7]. A well-balanced scheme is introduced later for balancing the flux gradients and source term approximation at each cell interface; see [1] and compare the results in [8,9]. Due to the simplicity of the rectangular cell, we implement a well-balanced scheme based on the weighted average flux (WAF) of the finite volume method extended from [8][9][10][11] to simulate various complex flow problems.…”

“…A well-balanced scheme is introduced later for balancing the flux gradients and source term approximation at each cell interface; see [1] and compare the results in [8,9]. Due to the simplicity of the rectangular cell, we implement a well-balanced scheme based on the weighted average flux (WAF) of the finite volume method extended from [8][9][10][11] to simulate various complex flow problems. The data elevation model (DEM) is available in rectangular format, so it is easy to utilize as an input in our program for simulating real-world problems.…”

“…One concept uses parallel computation. Our main study is to develop parallel programs on GPU where the parallel version is developed from the serial version proposed in [8,9]. The most challenging is the implementation using unstructured meshes since the data structure and hence the memory access is not straightforward.…”

“…Furthermore, this approach can be implemented on several platforms such as CPU [8,9,12,13], GPU [6][7][8][9][14][15][16][17], and a cluster of computers [3].…”

“…From these reasons, our paper focuses on optimizing a CUDA implementation of the shallow water equation [8,9]. Nsight, a CUDA profiler, is used to determine an area for computational improvement.…”