Overland-Gully-Sewer (2D-1D-1D) Urban Inundation Modeling based on Cellular Automata Framework

Abstract

A new overland-gully-sewer (2D-1D-1D) urban inundation modeling is herein proposed. 10 The authors have developed an integrated approach for overland-gully-sewer urban inundation simulations, which comprises 2D overland flow modeling (2D-OFM) by solving the two-dimensional non-inertia shallow water equations (2D-NISWEs), one-dimensional gully flow modeling (1D-GFM), and one-dimensional sewer flow modeling (1D-SFM). This approach can achieve good accuracy on urban inundation simulations. Nevertheless, such physically-based 2D overland flow modeling with the non-inertia SWEs (2D-OFM-NISWEs) requires substantial computational procedures, so that it is not suitable for real-time evaluation of inundation in urban areas. To resolve this limitation, the present study proposes a new integrated approach for overland-gully-sewer urban inundation modeling, in which the cellular automata (CA) framework is adopted for 2D-OFM to improve its suitability, and 1D-GFM and 1D-SFM are modeled by SWMM. A novel linkage among 2D-OFM-CA, 1D-GFM, and 1D-SFM is developed. A theoretical comparison of the governing equations and computational procedures between the two approaches is first conducted. Two historical flood events in Taipei city are next selected to perform the numerical accuracy verification and numerical efficiency evaluation. Analysis of inundation depths/velocities between the two approaches is conducted, and investigation of the drained and surcharged/inversed discharges of the two approaches is also made. The numerical comparisons between the two approaches indicate that the proposed approach can maintain good accuracy with significant reduction on its computing time. This newly proposed approach can finish a 3 hours short-duration urban inundation simulation on a case study with 335,621 cells within 12-17 mins. As a result, this CA-based approach is proved to be a useful tool for real-time urban inundation modeling.