High-Resolution Modelling With Bi-Dimensional Shallow Water Equations Based Codes-High-Resolution Topographic Data Use For Flood Hazard Assessment Over Urban And Industrial Environments-

摘要

The availability of new generation of High-Resolution (HR) topographic datasets combined with high performance computing resources opens the door to HR hydraulic simulations for risk assessment. LiDAR and photo-interpreted datasets are promising for HR Digital Elevation Model (DEM) generation, allowing inclusion of fine (infra-metric) aboveground structures influencing overland flow hydrodynamic in urban environment. Nonetheless, if topographic data is one key input for free surface hydraulic modelling using standard 2D Shallow Water Equations (SWEs) based codes, several categories of technical and numerical challenges arise to use HR dataset within numerical modelling. This proceeding explores the new possibilities, advantages and limits of HR topographic data use with 2D SWEs based numerical modelling tools for flood hazard assessment and proposes an original method for uncertainty assessment. The concepts of HR topographic data and 2D SWE based numerical modelling are reviewed. Using LiDAR and photo-interpreted datasets, different 2D SWEs based codes (Mike 21, Mike 21 FM, TELEMAC-2D, FullSWOF_2D) and strategies are tested to encompass HR DEM in intense rainfall and river flood events simulations ranging from industrial site scale to a megacity district scale (Nice, France). Tools and methods for assessing uncertainties related to HR DEM use with 2D SWE based codes are developed to perform a spatial global sensitivity analysis related to HR topographic data use. Computed sensitivity indices maps quantify the importance and spatial variability of uncertainties introduced by modeller choices regarding ways HR topographic information are integrated in models, compared to measurement errors. Impact of thin aboveground features inclusion, even at a decreased resolution, appears as a crucial asset in flood risk assessment on urban area, but requires providing caution to decision makers along with models' results.