Advancements in River-Floodplain Hydrodynamic Modelling for Urban Flood Hazard Assessment under Climate Change

Abstract

The diversity of flood‐generating mechanisms superimposed on catchment physiographic features with nonstationary meteorological drivers makes future urban flood hazard assessment a grand challenge. To date, many flooding predictions predominantly focus on hydrometeorological ensemble modelling chains, wherein river discharge is estimated at the catchment scale through the integration of ensemble climate simulations and hydrological models. These studies, considering changes to precipitation or river discharge, are informative regarding the direction and magnitude of changes, however, they do not provide the fine-scale spatial aspect of flood hazard that is important to understand future risks. Moreover, existing studies focusing on large-scale hydrodynamic flood hazard quantification often suffer from inadequately justified hydraulics. This limitation stems from the insufficient incorporation of river channels and the use of coarse grid resolutions, which fail to capture small-scale yet critical topographical features. Furthermore, the representation of hydrological processes is often poor due to the assumption of fully impervious surfaces in many hydrodynamic models. These factors collectively contribute to a fundamental gap in our understanding of the intricate interplay between climatic, hydrological, and hydrodynamic processes driving flood events.

This thesis aims to contribute to bridging this knowledge gap by advancing methods for physics-based river-floodplain hydrodynamic flood modelling, with a focus on enhancing reliability, computational efficiency, and the capacity to generate hydraulic information at an appropriate level of detail. The key contribution lies in integrating a novel 1D channel representation with a nonuniform structured 2D floodplain hydrodynamic model, along with improved representations of hydrological processes and GPU-based parallelization, facilitating large-scale yet locally relevant future urban flood modelling. Leveraging these advancements, this thesis explicitly considers both the evolving precipitation patterns, as identified in large ensemble convection-permitting climate models, and the dynamics of soil moisture, thus enhancing the evaluation of future urban flood hazards. Ultimately, these advancements improve flood hazard assessment strategies, facilitating better preparedness and response strategies for the evolving threat of climate change-induced floods.

Date of Award	1 Oct 2024
Original language	English
Awarding Institution	University of Bristol
Supervisor	Paul D Bates (Supervisor) & Jeff Neal (Supervisor)