An integrated approach to health risk assessment for sewer flooding

Abstract

Urban flooding has emerged as a pressing challenge, driven by many factors including rapid urbanisation, climate change, and inadequate ageing infrastructure. The introduction of pathogens, including bacteria, viruses, and other contaminants, into floodwaters amplifies public health concerns, emphasising the urgent need for effective mitigation and response strategies. Using numerical models offers researchers, flood risk analysts and urban planners a means to accurately simulate the linkage between sewer systems, surface flow paths and human health risks. However, there is still a relative lack of understanding and framework on how the microbial risk can be quantified from urban sewer flooding. A contemporary approach that is used to estimate infections from pathogens introduced from urban flood water lies within quantitative microbial risk assessments (QMRA). The results from flood and water quality models can be applied to QMRA outputs. This means that the spatial behaviour of flood extents, along with changes to pathogen concentrations and the associated risk to human health can be studied. However, it is clear that current approaches lack credible data for calibrating and validating numerical simulations. Such data includes flood depths, inundation times, flow velocities over a surface, known pipe flows and depths, sediment wash-off and water quality parameters. Collecting this data is difficult due to the harsh environment within urban flood water and sewer systems. In some cases, data acquisition may not be possible. Along similar lines modelling approaches have their own complications. These include the computational times of hydrodynamic simulations and model resolution affecting simulation accuracy. Other complications arise from accurately simulating sediment wash-off with realistic particle distributions. As a result, many flood modelling studies ignore sediment wash-off and its conveyance through sewer pipes. To enhance the understanding of flood modelling a 1D-2D hydrodynamic model in MIKE URBAN + and a sediment wash-off model in MIKE21 FM was developed. The models were calibrated and validated from unique datasets collected within physical models. To further improve our understanding of combined sewer flooding, this thesis proposes a methodology utilising outputs from several modelling approaches, including a 1D-2D hydrodynamic model (with pipe sediment scenarios), a stochastic sewer model, and a water quality model for QMRA applications. This thesis contributes to the integrated modelling approach to evaluate the health risk due to sewer flooding by demonstrating that calibrating a hydrodynamic model in a lower resolution (and therefore reduced simulation times) can still be applied to a more complex model of higher resolution. And the research proved that the accuracy of modelling graded sediment wash-off using a simplified uniform sediment model is still high. Therefore, both unique datasets can be used for future modelling studies. Both models (hydrodynamic in MIKE URBAN+ and sediment wash-off in MIKE21 FM) allow for the better understanding of drainage systems, which can help implement mitigation strategies